Nicotine Increases Size and Severity of Experimental Choroidal Neovascularization

Abstract

purpose. Cigarette smoking is the strongest environmental risk factor for all forms of age-related macular degeneration (AMD). In the present study, the influence of nicotine on the severity of choroidal neovascularization (CNV) in a mouse model of neovascular AMD and its effects on vascular smooth muscle cells derived from mouse choroid were investigated.

methods. A mouse model for CNV was used to study the effects of nicotine in young and middle-aged mice. Nicotine was administered orally in the drinking water to achieve serum levels consistent with those of chronic smokers. Hexamethonium, a nonspecific nicotinic receptor antagonist, was injected subconjunctivally to counteract the effects of nicotine. A mouse choroidal vascular smooth muscle cell line was exposed to nicotine, vascular endothelial growth factor (VEGF), platelet-derived growth factor (PDGF), or a combination of one of the factors and nicotine. Cell growth was determined by cell counts, and the activity of matrix metalloproteinase (MMP)-2 and -9 was quantified by gel zymography.

results. Nicotine administration resulted in increased size and vascularity of CNV, and older mice developed a greater relative increase than younger mice. This effect was blocked by subconjunctival hexamethonium. Choroidal vascular smooth muscle cells demonstrated a statistically significant increase in growth after exposure to a combination of PDGF and nicotine. Nicotine also reversed VEGF-induced suppression of MMP-2 activity.

conclusions. Nicotine increases size and severity of experimental CNV in the present mouse model, possibly by potentiating PDGF-mediated upregulation of proliferation of choroidal smooth muscle cells or by other mechanisms. These results suggest that non-neuronal nicotinic receptor activation probably mediates some of the harmful effects of cigarette smoking in wet AMD.

Age-related macular degeneration (AMD) is the leading cause of legal blindness in people 65 years of age or older in developed countries. ¹ The disease is classified into two clinical types: nonexudative, or dry, and neovascular, or wet. ² The most severe cases of vision loss are caused by neovascular AMD, due to the growth of abnormal new vessels under the retinal pigment epithelium (RPE) from the subjacent choroid, termed choroidal neovascularization (CNV). The pathogenesis of neovascular AMD is clearly multifactorial, with age, systemic health, genetic, and environmental risk factors playing roles in onset and progression. ¹ However, the pathogenic mechanisms whereby systemic health factors contribute to the regulation of CNV severity or induction remain unknown. 

Cigarette smoking is the environmental and systemic health factor with the greatest risk for onset and severity of all forms of AMD. ³ It has been associated with a two- to fourfold increased incidence of neovascular AMD. ^{3 4 5 6 7 8} The pathogenic mechanisms and specific factors in cigarette smoke that account for this increased risk are unknown. 

Cigarette smoke is a complex mixture of more than 4000 chemical substances that are distributed in particulate and gaseous phases. ^{9 10} The major components of the particulate phase are tar and nicotine, whereas the gaseous phase is composed primarily of carbon monoxide, carbon dioxide, and nitric oxide. Potential mechanisms by which cigarette smoke may cause end-organ damage include direct effects from chemicals in the smoke, immune activation, secondary hypoxia from pulmonary damage, and secondary sequelae from smoking-induced vascular disease. 

Recently, nicotine has been shown to contribute to many aspects of toxicity from cigarette smoke. ^{11 12 13} In the present study, nicotine increased the size and severity of CNV in a mouse model. The effect was blocked with a nonspecific antagonist of the nicotinic acetylcholine receptor in the same mouse model, and nicotine had a synergistic effect with PDGF in promoting proliferation of choroidal smooth muscle cells. 

Methods

Mice

Mice used in this study were handled in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Eight female C57BL/6 mice (National Institute of Aging, Bethesda, MD) were used for each group. 

Experimental Protocol

Animal studies were divided into two protocols:

1.  
   To evaluate nicotine’s effects in young (4 months) and middle-aged (11 months) female mice, one control (n = 8) and one nicotine-exposed (n = 8) subgroup for each age group was created.
2.  
   To evaluate the effect of nicotine receptor blockage with hexamethonium chloride, three groups (n = 8 each) of 11-month-old female mice were used. One control group received hexamethonium chloride (subconjunctival administration), and another group received nicotine by oral intake. A third group received both nicotine by oral administration and hexamethonium chloride by subconjunctival administration.

Laser Treatment

Laser-induced CNV was performed as previously described. ¹⁴ Briefly, a diode red laser beam was delivered to create four laser burns around the optic nerve of both eyes of the animals. Four weeks later, eyes were examined for CNV. 

Nicotine Administration and Measurement

Mice received nicotine (product number: N-1019; Sigma-Aldrich, St. Louis, MO) in their drinking water (100 μg/mL) for 4 weeks after laser treatment to produce serum levels similar to those of chronic, moderate smokers. ¹² Control mice received unaltered drinking water. Cotinine levels were measured at the time of death with an enzyme immunoassay (Cotinine MicroPlate, catalog no. 1124EA; OraSure Technologies, Inc., Bethlehem, PA). 

Hexamethonium Administration

Mice received daily subconjunctival injections of hexamethonium chloride (product number: H 2138; Sigma-Aldrich) at a concentration of 2.73 μg/μL with a special 30-gauge needle, syringe, and 2-μL repeating dispenser (Hamilton Co., Reno, NV) while under anesthesia. This drug treatment was started the same day the laser lesions were made, and continued for 2 weeks. 

Measurement of Lesion Severity

Fluorescein angiography, histopathology, and flatmount preparation and analysis were performed as has been described. ¹⁴  

Choroidal Vascular Smooth Muscle Cell Line

For primary cultures, mouse choroidal vascular smooth muscle cells (m-CVSMC) were isolated from C57BL/6 female mouse eyes. After enucleation, the eyes were rinsed with 10% gentamicin for sterilization and twice with PBS (1×). The eyes were then placed in a dish containing PBS (1×) and, with the aid of a dissecting microscope, opened by circumferential incision at the ora serrata. The anterior segment was removed, and the vitreous/retina was separated from the RPE and choroid eye cup with a round-tipped disposable blade (K20-1504) and Tennant forceps (K5-5230; both from Katena Products, Inc., Denville, NY). The RPE monolayer was dissected from Bruch’s membrane and choroid, under a dissecting microscope. The choroid was separated from the sclera with a Barraquer spatula (K3-2310; Katena Products, Inc.), aspirated with a micropipette, and transferred into a well coated with collagen IV (Sigma-Aldrich) and laminin (Invitrogen-Gibco, Grand Island, NY). Subsequently, cells were cultured, propagated, and maintained in Ham’s F-12 nutrient mixture supplemented with 15% fetal bovine serum (FBS, 1 mM l-glutamine, 100 μg/mL penicillin-streptomycin, and 0.075% sodium bicarbonate solution (all from Invitrogen-Gibco). Cells were cultured in 5% CO₂ at 37°C, the medium was changed every 3 days, and cells were passaged 1:2 when confluent. Cells at passages 4 to 7 were used for experiments. m-CSMC homogeneity was characterized by immunoreactivity to antibodies for actin and α-smooth muscle (Sigma). To ensure that there was no contamination with endothelial and epithelial cells, cells were stained with mouse anti-cytokeratin-18 antibody (Sigma-Aldrich) and with mouse anti-endothelial cell (CD146) monoclonal antibody (Chemicon International, Inc., Temecula, CA). 

Exposure to Nicotine, Vascular Endothelial Growth Factor, and Platelet-Derived Growth Factor-β and Cell Viability Studies

To determine the cellular responses to nicotine under various conditions, choroidal VSMC cells were plated on 96-well or 24-well plates coated with collagen IV/laminin with a density of 20,000 or 100,000 cells/well, respectively, and fed with regular medium supplemented with 10% phenol red-free FBS for 12 hours. Subsequently, the medium was changed to regular medium supplemented with 1% FBS for 24 hours. Cells were incubated in phenol red-free medium containing 0.1% fetal bovine serum with or without nicotine (Sigma-Aldrich) at physiologic concentrations comparable to those of active and passive smokers (1.0 and 0.01 μM, respectively) for a day. Then, the cells were incubated for 24 hours with or without 10 ng of VEGF (R&D Systems, Inc., Minneapolis, MN) or PDGF-B (Sigma-Aldrich) alone or in combination with nicotine. For each treatment, cell growth was determined by automated counting (Coulter Z1 cell counter; Beckman Coulter, Hialeah, FL) and by MTS assay (CellTiter 96 Aqueous One Solution Cell Proliferation Assay kit; Promega, Madison, WI). The supernatants were collected to assess activity of matrix metalloproteinase (MMP)-2 and -9. 

Quantification of MMP-2 and -9 Activity

Cell supernatants were collected 24 hours after treatment. At the time of medium collection, the cells at comparable density were counted for the purpose of adjusting the volume of the medium to the cell number. MMP-2 and -9 activity was assessed as described previously. ¹⁵ Standards were electrophoresed in parallel. Gels were incubated for 18 hours for MMP-2 and 48 hours for MMP-9 in 50 mM of Tris buffer. The gels were stained with Coomassie Blue and air dried. Densitometry was performed using ImageJ 1.17 (available by ftp at zippy.nimh.nih.gov/ or at http://rsb.info.nih.gov/nih-image; developed by Wayne Rasband, National Institutes of Health, Bethesda, MD), to determine relative MMP-2 and -9 activity. Each zymography assay was repeated three times. 

Statistical Analysis

Morphometric data for different lesions in each eye were averaged to provide one value per eye. The mean ± SD for these measures for each group was calculated, and probabilities (t-test) were calculated on computer (Prism 3.0; GraphPad, San Diego, CA). P < 0.05 was considered statistically significant for all forms of statistical analysis used. 

Data from statistical analysis of cellular responses and zymographies are expressed as a percentage of control or as arbitrary densitometry units. One-way analysis of variance and Dunnett’s multiple comparison post hoc was performed for the statistical analysis. 

Results

First, we confirmed that oral intake of nicotine achieved high blood levels by measuring serum cotinine, the major metabolite of nicotine. As expected, 4-month- and 11-month-old mice fed nicotine in water demonstrated similar serum cotinine levels (54.7 and 51.0 mg/dL, respectively). Control mice demonstrated unmeasurable levels. No evidence indicated that older mice had different levels than did younger mice. 

Next, we evaluated the severity of CNV in young (4-month-old) mice exposed to oral nicotine. Treated mice showed development of significantly larger CNV by both flatmount surface area and histopathologic measurement of thickness and diameter (Fig. 1) . Comparisons of fluorescein angiograms between nicotine-treated and control mice revealed that treated mice demonstrated larger areas of early hyperfluorescence and larger areas and intensities of late fluorescein leakage, indicating that the choroidal neovascular formation was more severe in treated animals (Fig. 2) . However, the frequency of individual lesions with evident fluorescein leakage was equivalent in both groups, indicating that nicotine treatment enhanced the severity of CNV. These results corroborated the flatmount analysis. 

We have shown that aging increases the severity of CNV. ¹⁴ We compared the response to nicotine of middle-aged (11-month-old) mice with that of young mice. As in the younger mice, nicotine-treated mice demonstrated greater size, surface area, and thickness than untreated, age-matched control mice, both by flatmount analysis and histopathology (Fig. 3) . Comparison of 4- to 11-month-old untreated control subjects revealed a significant increase in size in the older mice (cellular area: 1.3 ± 0.29 disc areas [DA] vs. 1.9 ± 0.30 DA, P < 0.002), which was consistent with our previous studies that demonstrated an influence of older age in CNV size and severity. In addition, the age-related difference in severity was observed after nicotine treatment, confirming an age-related increase in response to nicotine (cellular area: 2.1 ± 0.43 DA vs. 2.8 ± 0.79 DA, P < 0.044). 

To confirm that nicotine was acting through a nicotine receptor-dependent mechanism and not by an indirect effect of nicotine on a systemic factor (i.e., heart rate, blood flow), we evaluated the effect of subconjunctival administration of hexamethonium, a nicotinic receptor antagonist (Fig. 4) . As shown, hexamethonium treatment reduced the severity of CNV to control levels (Fig. 5) . 

Nicotine also increased the cellularity of CNV, as demonstrated by an increase in the cellular density index, reflecting a greater cellularity of CNV lesions in nicotine-treated animals, independent of their age (Fig. 6) . Because CNV are fibrovascular complexes comprising many cell types, including ECs, VSMCs, and others, this finding suggests that nicotine may act to increase growth of various cell types within CNV. 

Finally, we performed preliminary experiments to evaluate a potential mechanism by which nicotine may induce more severe CNV. Nicotine has been reported to contribute to vascular disease by its mitogenic activity on VSMCs, and by potentiating the proliferation induced by other growth factors. ^{16 17} Because laser-induced CNV contains abundant VSMCs, we evaluated the actions of nicotine on cultured mouse choroidal VSMCs, with or without the addition of VEGF or PDGF, two growth factors implicated in angiogenesis and vascular disease. Our data indicate that choroidal VSMCs demonstrated a dose-dependent increase in growth (i.e., increase in number of cells or MTS-assay response) in response to both VEGF and PDGF. VEGF and PDGF (10 ng each) were subthreshold concentrations (not shown). Addition of nicotine alone, in concentrations reflecting those observed in serum of active or passive smokers, did not increase choroidal growth of VSMCs (Fig. 7) . However, the addition of nicotine and subthreshold concentrations of PDGF, but not VEGF, significantly increased growth (Fig. 7) . Addition of nicotine with higher concentrations of VEGF or PDGF did not significantly alter cell growth over that observed with subthreshold concentration (not shown). 

The overexpression of MMPs by VSMCs is proposed as another major mechanism contributing to increased severity of pathologic neovascularization in many tissues. ¹⁸ We evaluated the effect of nicotine on the production of MMP-2 and -9, two gelatinases thought to be important in the formation of CNV. ¹⁹ Neither nicotine alone, PDGF alone, nor the combination of both changed MMP-2 activity released into the media by choroidal VSMCs (Fig. 8) . However, addition of VEGF alone partially suppressed MMP-2 activity under baseline conditions. This inhibitory effect was reversed in a dose-dependent manner by the addition of nicotine. The data for MMP-9 were similar (not shown). 

Discussion

In cross-sectional, case-control, and cohort studies, cigarette smoking has been shown to be the most important environmental risk factor for AMD. ¹ Among 4000 toxic compounds in cigarette smoke, nicotine is one of the most potent modulators of pathologic responses to vascular injury. ^{9 10} In particular, nicotine has been shown to be mitogenic for various types of endothelial cells and smooth muscle cells, to reduce apoptosis of vascular endothelial cells, and to induce the formation of capillary-like networks. ^{11 12 13 17} Furthermore, it has been demonstrated to stimulate angiogenesis, to accelerate the growth of atherosclerotic plaques, and to restore blood flow to ischemic limbs in mouse models. ¹² Therefore, nicotine is a good candidate for a chemical mediator in cigarette smoke to influence formation of CNV. 

Our results indicate that nicotine increases the severity of experimental CNV in a mouse model, and older mice are more affected than younger mice. ¹⁴ The effects of nicotine on CNV lesions was reversed with concurrent subconjunctival administration of hexamethonium, a nicotinic receptor antagonist. However, this route of administration may not have provided complete local receptor blockade, and it may also have had systemic effects on the size of the CNV lesion. Second, in vitro data suggest that nicotine potentiates choroidal VSMC growth in response to PDGF, and prevents VEGF-induced suppression of MMP-2 activity. ^{16 17 20 21 22 23 24} Taken together, these results suggest that smoking-related intake of nicotine may influence CNV formation by potentiating angiogenic activities, such as proliferation or MMP expression, induced by threshold concentrations of angiogenic or growth factors in vivo. 

Nicotine exerts its effects through heterodimeric receptors comprising α- and β-subunits. ^{25 26} These receptors are abundantly expressed within synapses of the central and peripheral nervous system. They are also widely expressed in peripheral non-neural cells, including hepatocytes, VSMCs, and ECs. ^{17 22 27 28} The tissue-specific differences in heterodimer combinations and expression have been studied in many tissues, but not in the choroid or retina. ^{26 27 29 30 31 32 33}  

The function, ligands, and signaling cascades of extraneuronal nicotinic receptors are not well understood. ^{22 26 34} For example, because nature is unlikely to have evolved non-neural peripheral nicotinic receptors without a purpose, the natural ligand for activation is not known. Although major blood vessels may have cholinergic innervation, capillary beds and neovascular tissue are unlikely to be innervated. Perhaps acetylcholine is locally synthesized in non-neuronal cells, or another unrecognized natural ligand may exist. Furthermore, signaling pathways for extraneuronal nicotinic receptors are unclear. ³⁵ Nicotine receptors on hepatocytes have been shown to induce a calcium response, ²⁸ but other signaling activities (i.e., phosphorylation) have not been examined. Nevertheless, although VEGF and PDGF use typical tyrosine kinase–mediated receptor signaling, ^{36 37 38} both molecules also activate calcium-dependent signaling pathways. ^{26 39 40 41} Thus, increased mobilization of calcium induced by nicotine could be a mechanism by which it potentiates the effect of these angiogenic stimuli, and calcium channel blockade might represent a potential inhibitory pathway. 

We also observed an age-dependent effect of responsiveness to nicotine. Research investigating the vascular biology of aging in nonocular vascular beds has suggested that blood vessels in old animals, compared with young blood vessels, often demonstrated abnormal pathologic responses to vascular injury. ^{42 43 44 45} It is surprising that no research has evaluated the differences between young and old responses to CNV. The specific age-related biological factors that predispose older mice to more severe CNV are unknown. However, increased severity of CNV is, in part, mediated by the age-related induction of abnormal over-responsiveness by the vascular cells in the CNV complex. Recently, our laboratory performed a very simple experiment comparing the size and severity of CNV in young versus old mice, in a laser-induced experimental CNV model. Our results demonstrated that CNV lesions in old animals were three times larger in surface area than CNV in young animals, and demonstrated greater thickness, fibrosis, and leakage. ¹⁴  

To our knowledge, this study is the first to evaluate in vitro responses of choroidal VSMCs to stimuli relevant to CNV formation. Although the importance of endothelial cells in CNV is self-evident, VSMCs are an important but often overlooked component of pathologic neovascularization and other vasculopathies. In other disorders, VSMCs contribute to lesion fibrosis, angiogenesis, and vascular remodeling. ⁴⁶ VSMCs have been observed in human CNV, but have not been carefully characterized. ⁴⁷ Recently, we have shown that approximately 40% of the cells within laser-induced CNV are VSMCs, suggesting an important role in this model. (Espinosa-Heidmann DG, et al. IOVS 2003;44:ARVO E-Abstract 3936) ⁴⁸ In ongoing experiments, we are evaluating the effect of cigarette-smoking–related stimuli on both VSMC and choroidal endothelial cells. 

Future studies are needed to characterize further the effects of whole cigarette smoke on CNV and to compare the results with the effects of nicotine. Also, the characterization and localization of specific nicotinic receptor subtypes on choroidal endothelial cells, on VSMCs, and within CNV are necessary. Ongoing experiments are also attempting to determine the mechanism for age-related increased severity. We hope that blockade of nicotinic receptors may have potential therapeutic implications as an adjunct to VEGF-blockade, which is already under investigation as a potential treatment for CNV. 

³  Contributed equally to the work and therefore should be considered equivalent senior authors.

 Presented in part at the annual meeting of the Association for Research in Vision and Ophthalmology, Ft. Lauderdale, Florida, May 2003.

 Supported by Veterans Affairs Merit Review Grant and National Eye Institute Grant EY/AI 13318.

 Submitted for publication July 14, 2003; revised August 25 and September 12, 2003; accepted September 16, 2003.

 Disclosure: I.J. Suñer, None; D.G. Espinosa-Heidmann, None; M.E. Marin-Castano, None; E.P. Hernandez, None; S. Pereira-Simon, None; S.W. Cousins, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Quantitative analysis of flatmount specimens and histopathology sections of the posterior pole of 4-month-old (young) mouse eyes 4 weeks after laser treatment. Top: vascular margins and cellular margin sizes (in disc areas, DA) of CNV lesions were significantly larger in nicotine-treated mice (1.543 ± 0.49 DA vs. 0.884 ± 0.23 DA, P < 0.0124; and 2.096 ± 0.44 DA vs. 1.292 ± 0.29 DA, P < 0.0028, respectively). Bottom: thickness in histopathology sections significantly increased in treated mice (9.77 ± 1.53 pixels vs. 7.52 ± 1.54 pixels, P < 0.0187), whereas the maximum diameter showed a trend toward a greater size in treated animals, but there was no statistically significant difference (P = 0.06). *Statistically significant difference.

Figure 2.

 

Fluorescein angiograms of eyes obtained 4 weeks after diode laser photocoagulation to induce CNV. Top: angiogram of a control animal. (A) A pointlike, faint hyperfluorescent area that moderately increases in size and intensity, as shown in frames (B) and (C). There was leakage in the late phases—the 1- and 5-minute frames, respectively, corresponding to intermediate (B) and late (C) phases—of the angiogram, which shows the relative severity of the CNV lesion. Bottom: fluorescein angiogram of a treated animal. A similar angiographic pattern can be seen, but the intensity and the size of the hyperfluorescent lesion was greater. (D) Initial phase of the angiogram in which a small area of hyperfluorescence was evident that grew intensively during the intermediate (E) and late (F) phases of the angiogram.

Figure 3.

 

Quantitative analysis of flatmount specimens and histopathology sections of the posterior pole of 11-month-old (middle age) mouse eyes 4 weeks after laser treatment. Top: vascular margin and cellular margin sizes in (disc areas, DA) of CNV lesions were significantly larger in nicotine-treated than in control mice (2.196 ± 0.95 DA vs. 1.426 ± 0.31 DA, P < 0.0301; 2.845 ± 0.79 DA vs. 1.949 ± 0.30 DA, P < 0.0148 respectively). Bottom: the maximum diameter and thickness in histopathology sections were significantly increased in treated mice (95.35 ± 36.19 pixels vs. 57.89 ± 14.02 pixels, P < 0.0194; 11.46 ± 3.64 pixels vs. 8.02 ± 2.23 pixels, P < 0.0485, respectively). *Denotes statistically significant difference.

Figure 4.

 

Flatmount preparation (propidium iodide stain) of the posterior pole of 11-month-old mouse eyes 4 weeks after laser treatment. (A) Eye of a control animal that received hexamethonium chloride subconjunctivally. CNV lesions were small and circular with discrete borders (dotted lines). (B) Eye of a mouse treated with nicotine in which there was coalescence of three laser injuries giving rise to a large CNV complex (dotted lines). (C) Eye treated with nicotine and hexamethonium. A reduction of the severity of CNV lesions can be clearly appreciated after blockage with hexamethonium. CNV lesions (dotted line) were similar in size to the control. Flatmount preparations provided an adequate visualization of the increased size and cellularity of the nicotine-treated animals (see quantitative analysis, Fig. 5 ). D, optic disc

Figure 5.

 

Quantitative analysis of flatmount specimens of 11-month-old mice receiving hexamethonium (H), nicotine (N), and hexamethonium + nicotine (H+N). All the parameters measured for CNV—vascular (A) and cellular margins (C), relative vascularity (B), and cell density (D)—were more severe in the nicotine-treated animals. *Statistically significant difference (P < 0.05).

Figure 6.

 

Quantitative analysis of flatmount specimens of the posterior pole of (left) 4- and (right) 11-month-old mouse eyes 4 weeks after laser treatment. CNV lesions in treated animals presented a higher index of cell density (1.985 ± 0.48 units vs. 1.397 ± 0.09 units, P < 0.0075; and 1.637 ± 0.19 units vs. 1.427 ± 0.06 units, P < 0.0142 respectively). *Statistically significant difference.

Figure 7.

 

Growth of m-CVSMCs determined by cell count. CVSMCs were exposed to various concentrations of nicotine (1.0–0.01 μM) for 24 hours; incubated for 24 hours, with or without 10 ng VEGF (A) or PDGF-β (B) alone or in combination with nicotine; and counted. Data are expressed as a percentage of control. Shown are mean ± SEM of three independent experiments run in triplicate on cultured cells. *Statistically significant difference (P < 0.01).

Figure 8.

 

m-CVSMC-derived MMP-2 activity evaluated by zymography in presence of nicotine (1.0–0.01 μM) alone and 10 ng VEGF (A) or PDGF-β (B), alone or in combination with nicotine. Top: gelatin zymogram from a representative experiment; bottom: averages of results of three independent experiments run in triplicate on cultured cells. *Statistically significant difference (P < 0.01) compared with control.