Molecular Vision 1999; 5:8 <>
Received 22 February 1999 | Accepted 17 May 1999 | Published 15 June 1999

The Rabbit Lens Epithelial Cell Line N/N1003A Requires 12-Lipoxygenase Activity for DNA Synthesis in Response to EGF

Mohammad S. R. Haque,1 Jaspreet K. Arora,1 George Dikdan,2 Thomas W. Lysz,2 Peggy S. Zelenka1

1Laboratory of Molecular and Developmental Biology, National Eye Institute, Bethesda, MD, 20892; 2Department of Surgery, Anatomy, and Cell Biology, UMDNJ, New Jersey Medical School, Newark, NJ 07103

Correspondence to: Peggy S. Zelenka, NIH/NEI/LMDB, Bldg 6/ Rm 214, 6 Center Drive, MSC 2730, Bethesda, MD 20892-2730; Phone: (301) 496-7490; FAX: (301) 435-7682; email:
Dr. Arora is now at Automation Group, QIAGEN, Inc., 28159 Avenue Stanford, Valencia, CA, 91355


Purpose: Cultured rat lenses and primary human lens epithelial cells (HLECs) express12-lipoxygenase (12-LOX) and require a 12-LOX metabolite of arachidonic acid for growth in response to EGF and insulin. This study seeks to identify an established cell line with these characteristics.

Methods: Immunoblotting was used to screen eight lens epithelial cell lines for 12-LOX expression: the human line, HLE-B3; mouse lines [alpha]TN4, 17EM15, 21EM15, and MLE6, and rabbit lines N/N1003A, LEP2 and B3. DNA synthesis was measured as incorporation of 3H-thymidine into DNA. Expression of c-fos mRNA was detected by RT-PCR. The involvement of 12-lipoxygenase metabolites was determined using the lipoxygenase inhibitors baicalein, cinnamyl 3,4-dihydroxy-[alpha]-cyanocinnamate (CDC), or nordihydroguiairetic acid (NDGA).

Results: 12-LOX was detected only in the rabbit lines N/N1003A, LEP2 and B3. N/N1003A cells were chosen for further study. 12-LOX inhibitors blocked DNA synthesis in response to EGF with or without insulin. Inhibition of EGF-stimulated DNA synthesis was reversed by 0.3 µM to 3 µM 12(S)hydroxyeicosatetraenoic acid (HETE), but not by equivalent concentrations of 5(S)HETE, 8(S)HETE, 15(S)HETE, or 12(R)HETE. Baicalein prevented EGF induction of c-fos mRNA. The transformed HLEC line, HLE-B3, showed little stimulation of DNA synthesis in response to EGF and was unaffected by the presence of 12-LOX inhibitors.

Conclusions: N/N1003A cells, like primary cultured human lens epithelial cells or neonatal rat lenses, require 12-LOX activity for EGF dependent growth. This line will be useful for studies of the mechanism of action of 12(S)HETE.


Epithelial cells isolated from human surgical specimens and from neonatal rat lenses have been shown to express 12-lipoxygenase (12-LOX), an enzyme responsible for synthesis of 12(S)HETE from arachidonic acid [1,2]. Moreover, inhibition of 12-LOX activity prevents organ-cultured neonatal rat lenses [2] or primary cultures of HLECs [1] from synthesizing DNA in response to EGF in the presence of insulin. Since exogenous 12(S)HETE reverses the effect of 12-LOX inhibitors on DNA synthesis, 12(S)HETE (or a further metabolite) seems to be required for lens epithelial cell growth in these culture systems. 12(S)HETE seems to exert its effect early in the mitogenic response, since cells treated with 12-LOX inhibitors fail to show induction of the immediate early gene, c-fos [1,2]. Interestingly, however, 12(S)HETE itself is not sufficient to cause DNA synthesis or c-fos induction in the absence of growth factors. These findings have suggested a cooperative interaction between 12(S)HETE and some component of the signal transduction cascade activated by the growth factors EGF and insulin.

A better understanding of the role of 12(S)HETE in lens epithelial cell growth is of fundamental interest because very little is known about the mechanism of action of this bioactive compound [3]. Moreover, further information about this pathway may have clinical applications for the treatment or prevention of posterior capsule opacification following cataract surgery [4]. However, more detailed studies of the mechanism of action of 12(S)HETE on lens epithelial cells have been hampered by the limited amounts of material available from dissected neonatal rat lenses or primary HLECs. Thus, the present study was undertaken to identify a lens epithelial cell line that expresses 12-LOX and shows the requirement for a 12-LOX metabolite of arachidonic acid previously observed in these other lens culture systems.


Cell Culture

We have examined 8 different lens epithelial cell lines: the human line, HLE-B3 (provided by Dr. Usha Andley, Washington Univ., MO); mouse lines [alpha]TN4 (provided by Dr. Paul Russell, NEI, Bethesda, MD), 17EM15, 21EM15, and MLE6 (provided by Dr. John Reddan, Oakland University, Rochester, MI), and rabbit lines N/N1003A, LEP2 and B3 (also provided by Dr. Reddan). Cells were cultured at 37 °C, 5% CO2. HLE-B3 cells were cultured in Minimal Essential Medium (MEM; Life Technology Inc., Gaithersburg, MD) supplemented with heat inactivated 20% fetal bovine serum (Life Technology Inc.), 100 U/ml penicillin and 100 µg/ml streptomycin. Mouse cell lines were plated with Dulbecco's Modified Eagle's Medium (DMEM; Life Technology Inc.) supplemented with 1 mM glutamine, heat inactivated 10% fetal bovine serum (Life Technology Inc.), 100 U/ml penicillin and 100 µg/ml streptomycin. Rabbit cell lines were cultured in Dulbecco's Modified Eagle's Medium (DMEM; Life Technology Inc.) supplemented with 1 mM glutamine, heat inactivated 10% rabbit serum (Life Technology Inc.), 100 µg/ml penicillin and 100 µg/ml streptomycin. All cells were plated at an initial density of 3 x 105/ml in 35 mm 6-well plates.

Inhibition of 12-lipoxygenase

For experiments involving inhibitors of 12-LOX, cells at 70-80% confluency were transferred to serum-free DMEM for 48 hours. At the end of this starvation period, fresh medium containing the inhibitors, growth factors and HETEs were added as indicated. The LOX inhibitor was added (30 µM baicalein [5], 30 µM NDGA [6] or 10 µM CDC [7]; Biomol, Plymouth Meeting, PA). After a 40 min pre-incubation period in the presence of the inhibitor, growth factors were added (15 ng/ml EGF, alone or with 1 µg/ml insulin; Life Technology Inc.). In certain experiments, HETEs dissolved in ethanol were added to the medium at final concentrations ranging from 0.1 µM to 3 µM (12(S)HETE, 5(S)HETE, 8(S)HETE, 15(S)HETE, or 12(R)HETE; Biomol). In experiments to test the ability of HETEs to reverse the effects of LOX inhibitors, the HETE and the LOX inhibitor were added simultaneously, 40 min before adding EGF (and insulin, when present).


The lens epithelial cells were washed with cold phosphate-buffered saline (PBS). Cells were lysed by adding PBSTDS (PBS, containing 1% Tritron X-100 [v/v], 0.5% [w/v] sodium deoxycholic acid, 1% SDS [w/v]), scraped off the plate, and transferred to a microcentrifuge tube on ice. The extract was sonicated in a cold water bath and the protein concentration was measured by the bicinchoninic acid assay [8] (BCA Protein Assay Reagent kit; Pierce, Rockford, IL). Aliquots of cell extract containing 20 µg protein were mixed with an equal volume of 2X loading buffer (Tris-Glycine-SDS Loading Buffer; Novex, San Diego, CA) and heated for 5 min at 100 °C. Electrophoresis was performed as previously described [9] in preformed 12% Tris-Glycine polyacrylamide gels (Novex, San Diego, CA) at 200 V for 45 min.

Following electrophoresis, the proteins were transferred to nitrocellulose membranes (0.45 µm pore size; Novex) for one hour at 100 V and immunoblotting was performed by a standard protocol [10]. Membranes were rinsed briefly in TBST (Tris-buffered saline with 0.05% Tween-20 (v/v)) and blocked overnight in 5% skim milk (Difco, Detroit, MI) in TBST. On the following day, the membranes were washed three times, 5 min each, in TBST. The membranes were incubated 2 h at room temperature in primary antibody (rabbit polyclonal anti-12-lipoxygenase; Oxford Biomedical Research, Oxford, MI, diluted to 1 µg/ml in TBST containing 2.5% skim milk). Membranes were washed three times for 5 min in TBST, then incubated on a shaker at low speed for 1 h at room temperature in the secondary antibody (goat anti-rabbit IgG linked to horseradish peroxidase, Biorad, Hercules, CA, diluted 1:3000 in 2.5% skim milk in TBST). Membranes were washed three times for 5 min in TBST and immunoreactive bands were detected with ECL+Plus (Amersham Pharmacia Biotech, England) using both autoradiography and fluorescence scanning with a Storm 860 phospho-fluoroimager (Molecular Dynamics, Sunnyvale, CA). Relative concentrations were quantitated using ImageQuant software (Molecular Dynamics).

Immunoassay of 12(S)HETE

For the assay of 12(S)HETE by enzyme-linked immunoassay (EIA; Perceptive Diagnostics, Cambridge, MA), the N/N1003A cells were serum deprived for 48 h and stimulated with EGF (15 ng/ml). The medium was collected after 24 h and assayed for 12(S)HETE by EIA as described previously [11]. The limit of detection was 0.17 ng/ml.

RNA Isolation

The N/N1003A lens epithelial cells were washed twice with cold PBS. The cells were lysed by adding 4 ml RNAzol (Tel-Test, Friendswod, TX) per 75 cm2 flask. RNA was extracted, precipitated, and washed following the manufacturer's protocol. The resulting pellet was dissolved in 50 µl diethylpyrocarbonate-treated water, and total RNA concentration was determined by absorbance at 260 nm. All RNA preparations had an A260/A280 ratio of 1.8 or higher.

Reverse Transcription-Polymerase Chain Reaction (RT-PCR)

For measurement of c-fos mRNA, cells were harvested 15, 30, or 120 min after stimulation with 15 ng/ml EGF. Where indicated, cells were pretreated 40 min with 30 µM of baicalein with or without 0.3 µM 12(S) HETE, prior to EGF treatment. Competitive RT-PCR with a DNA internal standard was used for quantitative assessment of c-fos mRNA with previously described oligonucleotides and assay conditions [2].

3H Thymidine Incorporation

For the measurement of DNA synthesis, N/N1003 or HLE-B3 cells were cultured in six well plates as described above. After adding EGF with or without insulin, the cells were incubated for 18 h, then labeled with 0.5 µCi/ml 3H-thymidine (sp. act. 44 Ci/mmol; Amersham Life Science, Arlington Heights, IL) in DMEM. After 1 h incubation, the labeled medium was removed. Cells were fixed in 1 ml of methanol-acetic acid (3:1 v/v) at room temperature for 10 min. The fixative was removed and cells were rinsed twice with 5 ml of 80% methanol. Finally 1 ml of 1% SDS was added to each well, the cells were scraped, and the entire lysate from each well was transferred to vials containing 9 ml of Biosafe scintillation fluid (Research Products International, Mount Prospect, IL) for measurement of radioactivity by scintillation counting.


Expression of 12-Lipoxygenase

We have examined eight lens epithelial cell lines for the presence of 12-LOX protein. The 12-LOX antibody detected a single band of the expected size (approximately 70 kDa) [12] only in the rabbit cell lines, N/N1003A, LEP2 and B3 (Figure 1A). In contrast, no immunoreactive protein was present in cell lines HLE-B3, 17EM15, 21EM15, MLE6 and [alpha]TN4 (Figure 1A). We immunoblotted adult mouse cornea as a positive control. The same 12-LOX antibody detected a single band of the expected size as N/N1003A cell line (Figure 1B) [13].

Synthesis of 12(S)HETE

To test whether N/N1003A cells synthesize 12(S)HETE, cells were cultured in the presence or absence of EGF for 24 h and the medium was analyzed for 12(S)HETE by an enzyme linked immunoassay. Cells cultured in the presence of EGF had a significantly higher concentration of 12(S)HETE in the medium (p<0.05) than cells cultured without EGF (Figure 2). The EGF dependent increase in 12(S)HETE concentration was significantly reduced (p<0.05) by the 12-LOX inhibitor, 10 µM CDC, suggesting that it is dependent on 12-LOX activity.

DNA synthesis

N/N1003A cells have a strong mitogenic response to EGF and insulin [10]. To determine whether N/N1003A cells show the same requirement for 12(S)HETE previously reported for the rat lens and primary human lens epithelial cells, we investigated the effect the 12-LOX inhibitor, baicalein, on the stimulation of DNA synthesis by EGF and insulin (Figure 3). Baicalein inhibited the incorporation of 3H-thymidine into DNA, with half-maximal inhibition of DNA synthesis observed at approximately 20 µM (data not shown). The inhibitory effect of baicalein on DNA synthesis was reversed by 12(S)HETE in a dose-dependent manner at concentrations between 0.1 µM and 2.0 µM (Figure 3). Addition of 10 µM CDC, a structurally unrelated 12-LOX inhibitor, produced a similar inhibition of DNA synthesis, which was also reversed by exogenous 12(S)HETE in this concentration range (data not shown). In the absence of EGF and insulin, comparable concentrations of 12(S)HETE had no effect on DNA synthesis (Figure 3).

Unlike the organ cultured rat lens and primary cultures of HLECs, which have been used previously to study the cellular response to 12(S)HETE, N/N1003A cells do not require the presence of insulin to synthesize DNA in response to EGF [14]. The optimal concentration of EGF for stimulation of DNA synthesis was approximately 15 ng/ml (data not shown). Since use of a single growth factor would simplify investigation of the mechanism of action of 12(S)HETE on lens epithelial cells, we tested whether DNA synthesis elicited by EGF in the absence of insulin retained the requirement for 12(S)HETE. The results demonstrated that DNA synthesis stimulated by EGF alone was strongly inhibited by a broad spectrum LOX inhibitor NDGA (30 µM, Figure 4) as well as by the more specific 12-LOX inhibitor CDC (10 µM, not shown). Exogenous 12(S)HETE completely reversed the inhibition of DNA synthesis by NDGA (Figure 4) or CDC (not shown). Other closely related HETEs, 5(S)HETE, 8(S)HETE, 15(S)HETE and 12(R)HETE, were unable to reverse the inhibitory effect of NDGA (Figure 5), demonstrating a specific requirement for 12(S)HETE or a metabolite of 12(S)HETE in the mitogenic response to EGF. Interestingly, the transformed human cell line, HLE-B3, which did not express 12-LOX, showed no significant stimulation of DNA synthesis in response to EGF and no inhibition of DNA synthesis by NDGA under identical experimental conditions (Figure 6).

Expression c-fos mRNA

Since expression of the immediate early gene, c-fos, occurs at or near the G0/G1 transition as cells enter the cell cycle, we examined the induction of c-fos mRNA in N/N1003A cells following stimulation by EGF. Addition of EGF to serum-starved cultures induced a transient increase in c-fos mRNA, which reached a maximum after about 15 min (Figure 7A). To examine whether products of the 12-LOX pathway are involved in the induction of c-fos mRNA expression, serum-starved cultures were pretreated with 30 µM baicalein for 40 min prior to addition of EGF. The presence of baicalein strongly inhibited the induction of c-fos mRNA (Figure 7B). Addition of 0.3 µM 12(S) HETE completely reversed the inhibitory effect of baicalein (Figure 7B).

Since these findings indicated that 12(S)HETE, or a metabolite of this compound, was required for induction of c-fos mRNA, we also tested whether 12(S)HETE alone was sufficient to induce c-fos mRNA in the absence of EGF or other growth factors. Serum starved cultures of N/N1003A cells were treated with 0.3 µM or 0.6 µM 12(S)HETE, concentrations that were able to reverse the inhibitory effect of baicalein on c-fos induction. In the absence of EGF, these concentrations of 12(S)HETE failed to induce c-fos mRNA expression (Figure 7B).


Examination of eight lens epithelial cell lines has identified three that express 12-LOX, an enzyme previously shown to be present in primary human lens epithelial cell culture and in neonatal rat lenses. Two related lipoxygenases are able to synthesize 12(S)HETE from arachidonic acid: platelet-type 12-LOX and leukocyte 12-LOX. Previous studies have established that platelet type 12-LOX is expressed in neonatal rat lenses [2] and HLECs [1]. In addition, recent work has shown that 15-LOX is expressed in lens fiber cells and may play an important role in organelle loss during lens fiber cell differentiation [15]. The antibody used to identify 12-lipoxygenase in the present study was raised against recombinant human platelet-type 12-lipoxygenase, but shows slight cross-reactivity with other lipoxygenases. Thus, although it is likely that the enzyme detected in the three rabbit lens epithelial cell lines is also the platelet-type 12-LOX, the immunoblotting results do not establish this with certainty.

The three lines that expressed 12-LOX were isolated as spontaneously immortalized clones from cultured rabbit lens epithelial cells and appear to be normal, non-transformed cells, as judged by morphology (J. Reddan, personal communication). Moreover, the N/N1003A line is diploid and does not form tumors in nude mice [16,17]. In contrast, two cell lines that were transformed with SV40 large T-antigen (human line HLE-B3 [18] and mouse line [alpha]TN4 [19]) contained no detectable levels of 12-LOX, suggesting that transformation of lens epithelial cells by T-antigen may suppress expression of this enzyme. Although the four remaining mouse cell lines were not intentionally transformed, their fibroblastic morphology suggests that they may have undergone spontaneous transformation (J. Reddan, personal communication).

Since the primary goal of this study was to identify a cell line in which the mechanism of action of 12(S)HETE could be investigated, we were pleased to discover that the N/N1003A cell line, which has already been well characterized with respect to growth factor requirements and gene expression [16,20], was among those that express 12-LOX. We therefore focused our attention on this line for further studies. Culturing N/N1003A cells in the presence of lipoxygenase inhibitors demonstrated that these cells have the same requirement for a 12-LOX metabolite of arachadonic acid as organ cultured neonatal rat lenses and primary cultures of human lens epithelial cells. Inhibiting 12-LOX activity with any of three structurally different lipoxygenase inhibitors (baicalein, CDC, and NDGA) profoundly inhibited incorporation of 3H-thymidine into DNA following stimulation by EGF with or without insulin. As in other studies on lens epithelial cells, the inhibitory effect of these compounds was exerted early in the cell cycle, since the EGF-stimulated N/N1003A cell showed no induction of the immediate early gene, c-fos. The effect of lipoxygenase inhibitors on DNA synthesis was reversed by adding exogenous 12(S)HETE. Exogenous 12(S)HETE was effective at concentrations 100-fold lower than the various inhibitors, ruling out direct binding of 12(S)HETE to the inhibitor or direct competition for a binding site as a possible mode of action.

The greater amount of material provided by the N/N1003A cells allowed us to extend observations that were initially made with neonatal rat lenses and primary cultures of HLECs. In particular, we explored the ability of a variety of structurally related HETE's to reverse the inhibition of DNA synthesis caused by the broad spectrum LOX inhibitor, NDGA. At a concentration of 0.3 µM, 12(S)HETE significantly reversed the inhibitory effect of NDGA. Four structurally related HETE's were unable to substitute for 12(S)HETE even at concentrations as high as 3.0 µM. These results demonstrate that the requirement for 12(S)HETE is highly specific and suggest that 12(S)HETE may exert its effect through a stereospecific interaction with a cellular protein.

Since N/N1003A cells grow in response to EGF alone, without a requirement for insulin, they provide a simpler system in which to analyze the mechanism of action of 12(S)HETE. In this cell line, DNA synthesis stimulated by EGF alone is sensitive to inhibition by lipoxygenase inhibitors. Moreover, the lipoxygenase inhibitors exert their effects early in the cell cycle, at or prior to induction of c-fos mRNA, whether or not insulin is present. Although exogenous 12(S)HETE reverses the effects of lipoxygenase inhibitors, it is unable to induce c-fos mRNA expression in the absence of EGF. This indicates that 12(S)HETE does not activate an alternative signaling pathway leading to c-fos transcription. Rather, the data suggest that 12(S)HETE (or a metabolite) interacts with some component of the EGF-stimulated signal transduction cascade upstream of c-fos transcription. Further studies of this cell line will focus on identifying the point of interaction.


We thank Drs. John Reddan, Paul Russell, and Usha Andley for providing the cell lines used in this study.


1. Arora JK, Lysz TW, Zelenka PS. A role for 12(S)-HETE in the response of human lens epithelial cells to epidermal growth factor and insulin. Invest Ophthalmol Vis Sci 1996; 37:1411-8.

2. Lysz TW, Arora JK, Lin C, Zelenka PS. 12(S)-hydroxyeicosatetraenoic acid regulates DNA synthesis and protooncogene expression induced by epidermal growth factor and insulin in rat lens epithelium. Cell Growth Differ 1994; 5:1069-79.

3. Honn KV, Tang DG, Gao X, Butovich IA, Liu B, Timar J, Hagmann W. 12-lipoxygenases and 12(S)-HETE: role in cancer metastasis. Cancer Metastasis Rev 1994; 13:365-96.

4. Beebe DC. Control of lens growth: relationship to secondary cataract. Acta Ophthalmol Suppl 1992; 205:53-7.

5. Sekiya K, Okuda H. Selective inhibition of platelet lipoxygenase by baicalein. Biochem Biophys Res Commun 1982; 105:1090-5.

6. Salari H, Braquet P, Borgeat P. Comparative effects of indomethacin, acetylenic acids, 15-HETE, nordihydroguaiaretic acid and BW755c on the metabolism of arachidonic acid in human leukocytes and platelets. Prostaglandins Leukot Med 1984; 13:53-60.

7. Cho H, Ueda M, Tamaoka M, Hamaguchi M, Aisaka K, Kiso Y, Inoue T, Ogino R, Tatsuoka T, Ishihara T, et al. Novel caffeic acid derivatives: extremely potent inhibitors of 12-lipoxygenase. J Med Chem 1991; 34:1503-5.

8. Smith PK, Krohn RI, Hermanson GT, Mallia AK, Gartner FH, Provenzano MD, Fujimoto EK, Goeke NM, Olson BJ, Klenk DC. Measurement of protein using bicinchoninic acid. Anal Biochem 1985; 150:76-85.

9. Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970; 227:680-5.

10. Towbin H, Staehelin T, Gordon J. Electrophoretic transfer of protein from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc Natl Acad Sci 1979; 76:4350-4.

11. Scioscia KA, Snyderman CH, Rueger R, Reddy J, D'Amico F, Comsa S, Collins B. Role of arachidonic acid metabolites in tumor growth inhibition by nonsteroidal antiinflammatory drugs. Am J Otolaryngol 1997; 18:1-8.

12. Chen XS, Brash AR, Funk CD. Purification and characterization of recombinant histidine-tagged human platelet 12-lipoxygenase expressed in a baculovirus/insect cell system. Eur J Biochem 1993; 214:845-52.

13. Hurst JS, Balazy M, Bazan HE, Bazan NG. The epithelium, endothelium, and stroma of the rabbit cornea generate 12(S)-hydroxyeicosatetraenoic acid as the main lipoxygenase metabolite in response to injury. J Biol Chem 1991; 266:6726-30.

14. Reddan JR, Wilson-Dziedzic D. Insulin growth factor and epidermal growth factor trigger mitosis in lenses cultured in a serum-free medium. Invest Ophthalmol Vis Sci 1983; 24:409-16.

15. van Leyen K, Duvoisin RM, Engelhardt H, Wiedmann M. A function for lipoxygenase in programmed organelle degradation. Nature 1998; 395:392-5.

16. Reddan JR, Chepelinsky AB, Dziedzic DC, Piatigorsky J, Golenberg EM. Retention of lens specificity in long-term cultures of diploid rabbit lens epithelial cells. Differentiation 1986; 33:168-74.

17. Meakin SO, Reddan JR, Tsui LC, Brietman ML. A rabbit lens epithelial cell line supports expression of an exogenous crystallin gene characteristic of lens fiber cell differentiation. Exp Eye Res 1989; 48:131-7.

18. Andley UP, Rhim JS, Chylack LT Jr, Fleming TP. Propagation and immortalization of human lens epithelial cells in culture. Invest Ophthalmol Vis Sci 1994; 35:3094-102.

19. Yamada T, Nakamura T, Westphal H, Russell P. Synthesis of alpha-crystallin by a cell line derived from the lens of a trangenic animal. Curr Eye Res 1990; 9:31-7.

20. Sax CM, Dziedzic DC, Piatigorsky J, Reddan JR. Analysis of alpha-crystallin expression in cultured mouse and rabbit lens cells. Exp Eye Res 1995; 61:125-7.

Haque, Mol Vis 1999; 5:8 <>
©1999 Molecular Vision <>
ISSN 1090-0535