Expression, Localization, and Correlation of N-Myristoyltransferase and Its Inhibitor in Bovine Eye

Abstract

purpose. N-myristoyltransferase (NMT) is a ubiquitously distributed eukaryotic enzyme that catalyzes myristoylation of proteins. Very little is known about the process of myristoylation, particularly in the eye. In the present study, the distribution, expression, and correlation of NMT and its inhibitor (NMT inhibitor protein, NIP) in the bovine eye were investigated.

methods. Whole bovine eyes were either homogenized or regionally dissected to determine the activity and localization of NMT and NIP. Dissected tissues were homogenized, and Western blot analysis was performed using polyclonal anti-NMT and anti-NIP antibodies. The NMT activity was assayed using cAMP-dependent protein kinase or pp60^src derived peptide as a substrate. Fresh samples were then prepared for immunohistochemical analysis and probed with polyclonal anti-NMT and anti-NIP antibodies.

results. The total bovine eye cytosolic fraction displayed both NMT and NIP expression. NMT was present in all the regions of the eye at various levels of expression. The highest expression of NMT was in the cornea, whereas NIP was present in the retina, optic nerve, sclera, and choroid only. NIP expression was the highest in the optic nerve, sclera, and retina. NMT activity was observed in the cornea, iris, and retina after DEAE-Sepharose CL-6B column chromatography. The inhibitory activity of crude homogenate on recombinant human NMT activity was found to be greater for optic nerve and choroid. Immunohistochemistry results displayed similar findings.

conclusions. The varied expression of NMT in different regions of the eye reveals a regulatory relationship of NMT with NIP. These findings indicate that NMT and NIP are present in various regions of the eye and will lead to further understanding of visual signaling in ocular cells.

Myristoylation is a cotranslational modification of proteins with myristic acid (a 14-carbon saturated fatty acid). This modification regulates both protein function and localization. ^{1 2} Myristoylated proteins are acylated through an amide linkage to their N-terminal glycine residue, a reaction catalyzed by N-myristoyltransferase (NMT). ^{2 3 4 5 6} Myristoylated proteins in the cell have diverse biological functions such as signal transduction, cellular transformation, and oncogenesis. ^{2 3 4 5 6} Myristoylated proteins include the catalytic subunit of cAMP-dependent protein kinase, ⁷ the β-subunit of calcineurin, ⁸ the α-subunit of several G-proteins, ⁹ cellular and transforming forms of pp60^src, ¹⁰ and several tyrosine kinases and proteins essential for the assembly, maturation, and infectivity of mature virus particles, such as the murine leukemia virus Pr65^gag precursor ¹¹ and poliovirus VPO polypeptide precursor. ¹² NMT knockout Drosophila resulted in various embryonic cytoskeletal and cell morphologic defects and apoptotic cell death, suggesting the significance of myristoylation. ¹³ During tumorigenesis, the importance of protein myristoylation was first suggested by the fact that myristoylation of the viral oncogene pp60^src is required for membrane association and cell transformation. ^{14 15} Elevated expression of NMT in colorectal and gallbladder cancer has been reported from our laboratory, ^{16 17} suggesting NMT could be a prognostic marker for cancer. 

N-myristoylation of protein is crucial for HIV-1 Gag, ^{18 19 20} which renders proper functions and survival. NMT inhibitors were found to prevent the binding of Gag to membrane and virus assembly. ²¹ Thus, NMT is considered to be one of the key proteins for both HIV-1 and its host cell. It has been shown that, with decreased expression levels of NMT, HIV structural proteins were expressed gradually in HIV-1 infection of the human T-cell line CEM. This is attributed to the virus’s strategy for persistent replication. Also, there is decrease in the mRNA levels of human NMT isoforms and the NMT activities in the course of infection. ^{22 23}  

Rhodopsin, a visual pigment in vertebrate rod photoreceptors, is activated by light, which leads to hydrolysis of cGMP and closure of the cGMP-gated channel to induce membrane hyperpolarization. ^{24 25} Recoverin is a calcium-binding protein that inhibits rhodopsin phosphorylation by rhodopsin kinase until rhodopsin is activated by light. ²⁶ The posttranslational modification of N-terminal glycine of recoverin is essential for the association of recoverin with rod segment membranes at high Ca²⁺ concentrations. ^{27 28} Recoverin is membrane bound in the dark adapted state and a decrease in bound Ca²⁺ induced by light causes conformational changes in recoverin. This results in binding of Ca²⁺ to recoverin and extrusion of the myristoyl group known as “myristoyl switch,” which is unique to recoverin. ²⁹  

In our continuing interest in protein myristoylation, we initiated an investigation to study the localization, expression, and activity of NMT in the eye. In view of the fact that the eye is a heterogeneous organ containing many different regions, and very little is known about the mechanism of NMT action, we report here for the first time the distribution of NMT in the different regions of eye and its regulatory relationship with NMT inhibitor protein (NIP). 

Methods

Materials

9,10-³H myristic acid (39.3 Ci/mmol) was purchased from Perkin Elmer Life Sciences (Boston, MA). Pseudomonas acyl CoA synthetase, phenylmethylsulfonyl fluoride, soybean trypsin inhibitor, pepsin benzamidine, and monoclonal anti-β-actin antibody were from Sigma-Aldrich (Toronto, Ontario, Canada). Peptide substrate based on the N-terminal ends of cAMP-dependent protein kinase (GNAAAAKKRR) and pp60^src (GSSKSKPKR) were synthesized by the Alberta Peptide Institute (Edmonton, Canada). Polyclonal antibodies were raised against purified human NMT and NIP in New Zealand White rabbits. The specificity of these antibodies has been described. ^{17 30} Recombinant human NMT was purified as described by Raju et al. ³¹ Nitrocellulose membranes and HRP-conjugated secondary antibody were from BioRad Laboratories (Hercules, CA). Chemiluminescence agents were from Perkin Elmer Life Sciences. Horseradish peroxidase (HRP)–conjugated secondary and anti-von Willebrand antibodies for immunohistochemistry were purchased from Dako Corp. (Carpinteria, CA). General laboratory reagents were of analytical grade and obtained from Sigma-Aldrich. 

Preparation of Homogenates

All methods conformed to the ARVO Statement for the Use of Animals in Ophthalmic and Visual Research. Fresh bovine eyes were obtained from an abattoir, transported on ice, and used immediately. Various bovine eye tissues (cornea, lens, iris, retina, vitreous body, optic nerve, sclera, and choroid) were dissected and homogenized (Polytron; Brinckman Instruments, Westbury, NY), five times for 30 seconds each in 2 mL ice-cold phosphate-buffered saline (PBS) containing 0.1 mM EGTA and 10 mM 2-mercaptoethanol. This was further centrifuged at 10,000g at 4°C for 30 minutes to obtain a clear homogenate. Total eye cytosol was prepared from the whole bovine eye with a meat grinder followed by homogenization in a blender for 1 minute in PBS containing 2 mM EDTA, 10 mM 2-mercaptoethanol, 100 mg/L phenylmethylsulfonyl fluoride, 100 mg/L soybean trypsin inhibitor, and 200 mg/L benzamidine. Centrifugation at 100,000g at 4°C for 1 hour yielded the cytosolic fraction. The crude homogenates of different regions of eye were applied to a DEAE-Sepharose CL-6B column (1.4 × 1 cm) pre-equilibrated with PBS containing 0.1 mM EGTA and 10 mM 2-mercaptoethanol. The column was subsequently washed with two bed volumes of the above-described buffer, and proteins were eluted with buffer containing 1 M NaCl. 

N-myristoyltransferase Assay

N-myristoyltransferase activity was assayed as described previously. ^{32 33} Briefly, [³H]-myristoyl-CoA was synthesized as described previously. ³³ The reaction mixture contained 40 mM Tris-HCl (pH 7.4), 0.1 mM EGTA, 10 mM MgCl₂, 5 mM adenosine triphosphate (ATP), 1 mM LiCoA, 1 μM [³H]myristic acid (7.5 μCi), and 0.3 U/mL Pseudomonas acyl-CoA synthetase in a total volume of 200 μL. The reaction was performed for 30 minutes at 30°C. The conversion to [³H]myristoyl-CoA was generally greater than 95%. The assay mixture contained freshly generated [³H]myristoyl-CoA, 225 mM Tris-HCl (pH 7.4), 0.5 mM EGTA, 0.45 mM 2-mercaptoethanol, 1% Triton X-100, 500 μM of pp60^src-derived peptide or cAMP-dependent protein kinase–derived peptide and crude homogenate of various regions of the eye as a source of NMT in a total volume of 25 μL. The transferase reaction was initiated by the addition of radiolabeled myristoyl-CoA and was incubated at 30°C for 30 minutes. One unit of NMT activity was expressed as 1 picomole of myristoyl peptide formed per minute. 

N-myristoyltransferase Inhibition Assay

NIP was assayed by its inhibitory activity against standard human NMT. Crude homogenates of various regions of the eye were assayed for NIP activity against purified human NMT (0.2 μg/assay) in the presence of cAMP-dependent protein kinase–derived peptide. A control experiment was performed in the absence of eye homogenates, and human NMT activity was considered to be 100%. All other conditions were as just described, except the assay buffer contained 40 mM Tris-HCl (pH 7.4). 

Sodium Dodecyl Sulfate–Polyacrylamide Gel Electrophoresis and Western Blot Analysis

SDS-PAGE (10%) was performed according to the method of Laemmli. ³⁴ Western blot analyses were performed essentially as described by Towbin et al. ³⁵ Briefly, equal amounts of proteins were subjected to 10% SDS-PAGE and transferred to a nitrocellulose membrane. The membrane was blocked with 5% powdered milk in PBS-Tween 20 (PBST) and polyclonal anti-human NMT or anti-NIP antibody or monoclonal anti-β-actin was added and incubated overnight at 4°C. After washing with PBST, the membrane was incubated with HRP-conjugated secondary goat anti-rabbit or anti-mouse antibody. Immunoreactive bands were visualized on x-ray film by using chemiluminescence reagents. 

Immunohistochemistry

Immunohistochemistry was performed as described previously. ³⁶ Bovine eyes were fixed in 10% formaldehyde, dehydrated, and embedded in paraffin. Briefly, the sections were deparaffinized, rehydrated, incubated with hydrogen peroxide (0.5% in methanol), treated with pepsin (2 mg/mL 0.01 N HCl) for 45 minutes, and blocked with 1% BSA in PBS for 30 minutes. Sections were incubated with anti-human NMT antibody (1:100) for 90 minutes and HRP-conjugated secondary antibodies (1:100) for 45 minutes followed by color development. Controls included omission of primary antibody or both primary and secondary antibodies or staining with anti-von Willebrand Factor antibody (dilution, 1:100). 

Other Methods

Protein concentrations were determined by the method of Bradford using BSA as a standard. ³⁷  

Results

N-Myristoyltransferase Activity in the Eye

To analyze the role of myristoylation in the bovine eye, we examined NMT activity in the crude homogenates of cornea, lens, iris, retina, vitreous body, optic nerve, sclera, and choroid, using pp60^src or cAMP-dependent protein kinase–derived peptide as a substrate. NMT activity was not observed in any region of the eye. However, when the crude homogenate was applied to a DEAE-Sepharose CL-6B column, the high-salt (1 M NaCl) eluent of the cornea, iris, and retina showed NMT activity (Fig. 1) . NMT activity was high in the presence of pp60^src-derived peptide compared with that in the presence of cAMP-dependent protein kinase–derived peptide (Fig. 1) . The observed NMT activity after DEAE-Sepharose chromatography suggests that an unidentified inhibitor(s) of NMT may have been removed. To test this possibility, the inhibition of standard human NMT was performed by homogenates of various regions of the eye. Retina, optic nerve, sclera, and choroid showed inhibitory effects on human NMT activity (Fig. 2) . The whole cytosolic fraction of eye was devoid of NMT activity, whereas it had an inhibitory effect on human NMT in vitro (data not shown). 

Differential Expression of NMT and NIP Proteins

The crude homogenate showed NMT activity only after DEAE-Sepharose chromatography suggesting partial removal of inhibitor protein(s). Consequently, we investigated the protein expression profile of NMT in different regions of the eye by Western blot analysis. Crude homogenates of different regions of the eye were probed with polyclonal antibodies raised against NMT. Western blot analysis of eye samples revealed immunoreactive bands for NMT at 58 kDa in all regions of the eye with various levels of protein expression (Fig. 3A) . The highest expression of NMT was observed in the cornea, and moderate expression was observed in the retina (Fig. 3A , lane 4 vs. 5) whereas the lens, vitreous body, optic nerve, sclera, and choroid showed low expression of NMT. Though NMT was expressed in all the regions of the eye, no enzymatic activity was observed in the crude homogenates. To explore the possibility of the presence of inhibitor(s) of NMT, we investigated the expression of NIP(s) in crude homogenates of different regions of the eye (Fig. 3B) . On probing with anti-NIP antibody, the retina, optic nerve, and sclera showed a major band at 45 kDa. High expression of NIP was observed in the retina, optic nerve, and sclera (Fig. 3B , lanes 5 to 7), whereas the choroid showed low expression. It is interesting to note that the retina and optic nerve showed multiple immunoreactive bands. The explanation for these multiple immunoreactive bands may be the proteolytic degradation of inhibitor protein(s). However, because the retina and optic nerve are heterogeneous tissues containing many different cell types, it is possible that the multiple immunoreactive bands originated from different cell types. We did not detect immunoreactive bands in other regions of the eye. The results suggest that protein(s) other than 45-kDa protein may inhibit NMT activity in crude homogenates of cornea, lens, and iris. One possible inhibitor may be serum albumin, which inhibits NMT activity, as we have shown. ³⁸ Furthermore, the total eye cytosolic fraction showed expression of both NMT and NIP (Figs. 3A 3B , lane 9). 

Immunohistochemistry

Further regarding our observation of the ubiquitous expression of NMT in all regions of the eye, we investigated the localization of NMT and NIP(s) in various regions of the eye by immunohistochemical methods. Sections from cornea, retina, and optic nerve incubated with only secondary antibody showed no staining, indicating the absence of nonspecific binding (Fig. 4A) . Incubation with anti-vWF antibody, however, outlined the vasculature in those sections (Fig. 4B) . Sections prepared from bovine eye and incubated with anti-human NMT antibody showed a distinct reaction in corneal epithelium, whereas the substantia propria was negative (Fig. 4C) . Moderate staining was observed for human NMT in the retina (Fig. 4D) , whereas the choroid (Fig. 4D) and optic nerve (Fig. 4E) did not show any staining. Furthermore, sections from the eye when incubated with anti-NIP antibody showed a distinct reaction with the retina (Fig. 4F) , optic nerve (Fig. 4G) , ciliary body, and lens (Fig. 4H) , whereas all other sections showed no reaction. 

Discussion

All the regions of the eye showed the presence of NMT. The expression of NMT varied within the tissue, with the highest expression in the cornea, iris, and retina. All purified NMT-1 essentially contain a single polypeptide with molecular mass within a range of 49 to 68 kDa. ^{39 40} The molecular mass and cross reactivity with polyclonal antibody raised against NMT-1 suggest that bovine eye NMT belongs to the NMT-1 family. The crude homogenates did not show any NMT activity due to the presence of NIP(s). Therefore, we applied crude homogenate to a DEAE-Sepharose CL-6B column to partially separate NIP(s). After chromatography, the crude homogenate of the cornea, iris, and retina showed NMT activity. Though the expression of NIP in the retina and optic nerve was similar, the inhibitory activity in optic nerve was found to be much higher than in the retina. The varied expression of NMT and NIP in various regions of the eye and selective NMT activity in cornea, iris, and retina suggests that NIP(s) regulates NMT activity in the eye. Therefore, NMT may play a significant role in the lens and retina. In the lens, a single protein of 19 kDa is myristoylated, which is thought to have significant physiological functions and is restricted to proliferating epithelial cells. ⁴¹ Retina is the warehouse of the phototransduction proteins such as recoverin and transducin (G_tα). It has been suggested that, because of the absence of NMT in photoreceptor cells, there is no myristoylation in G_tα derived from the retina. ⁴² Yang and Wensel ⁴³ reported that G_tα is myristoylated in the retina but is not sufficient for membrane association. Our results also support the presence of NMT activity in the retina. Furthermore, the insufficient membrane association of G_tα may be due to the high expression of NIP. The myristoylated and nonacylated forms of recombinant recoverin have been used to study the effect of myristoylation on membrane binding. The results revealed that only myristoylated recombinant recoverin showed Ca²⁺-induced binding to membranes. ^{29 44} N-terminal acylated proteins in photoreceptor cells include G_tα, GCAP, recoverin, and a catalytic subunit of cAMP-dependent protein kinase. Low or nonexistent NMT activity observed in different regions of the eye may be due to both NIP and the fact that there is heterogeneous acylation in the eye. The substitution of myristate by other fatty acids in recoverin has weak inhibitory effects on the rhodopsin phosphorylation ²⁶ and is primarily responsible for its localization. The fatty acids, 12:0, 14:0, 14:1n-9, and 14:2n-6 present in photoreceptor proteins ⁴⁵ are available to the NMT, and all of them compete for NMT. However, it has the highest affinity for 14:0 CoA. The present study suggests that NMT is directly regulated by NIP in addition to the heterogeneous acylation in retina. The expression and localization of NMT and NIP was further supported by immunohistochemistry results. 

Peroxisomal β-oxidation is found to be defective in eye diseases such as adrenomyeloneuropathy, infantile Refsum disease, and Zellweger syndrome. ^{46 47 48 49} N-terminal fatty acylation with 14:1n-9 and 14:2n-6 are retroconverted from 14:0 in the retina ⁵⁰ and is vital for normal functioning of phototransduction proteins. Any defect in their biosynthesis has deleterious effects on the normal functioning of the eye. Because there is high expression of NIP in the retina, it can play a crucial role in regulating NMT-mediated biosynthesis of these fatty acids. 

Our study demonstrates the first characterization and localization of NMT and NIP(s) in the various regions of eye. This investigation also suggests that NIP can be a target candidate for treatment of disease in the eye. These results will lead to further understanding of visual signaling in ocular cells. 

 Supported in part by Grant MOP-36484 from the Canadian Institutes of Health Research.

 Submitted for publication November 3, 2003; revised January 20, 2004; accepted February 5, 2004.

 Disclosure: A. Shrivastav, None; M.K. Pasha, None; P. Selvakumar, None; B. Singh, None; R.K. Sharma, 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.

 

NMT activity in different regions of the eye. Crude homogenates of different regions of the eye were applied to a DEAE-Sepharose CL-6B column and 1 M NaCl elute was assayed for NMT activity using pp60^src (▪) or cAMP-dependent protein kinase A (□)–derived peptide. Only cornea, iris, and retina displayed NMT activity. Data are expressed as the mean ± SD of results in three independent experiments.

Figure 2.

 

Inhibitory activity of various regions of the eye on recombinant human NMT. Inhibitory activity in various regions of the eye was assayed. Results are expressed as the percentage inhibition of human NMT activity considering 100% activity for human NMT in the absence of eye homogenates. Data are expressed as the mean ± SD of results in three independent experiments.

Figure 3.

 

Western blot analysis of eye tissue. Proteins (25 μg) from the various regions of bovine eye shown above the lanes were subjected to 10% SDS-PAGE, transblotted onto nitrocellulose membrane, and were probed with (A) polyclonal anti-human NMT antibody (1:1000), (B) polyclonal anti-NIP antibody (1:1000), (C) monoclonal anti-β actin antibody (1:1000, used as the loading control indicator).

Figure 4.

 

Immunohistochemicalanalysis of various regions of the eye. Eye sections incubated with only secondary antibody (A) lacked staining, whereas an adjacent section stained with anti-vWF antibody (B) showed reaction in blood vessels (arrows). Anti-NMT antibody reacted with epithelium (arrow) but not with substantia propria (S) in the cornea (C; Co). (D) NMT localization in retina (R) but not choroid (Ch). Optic nerve (E, ON) was negative for NMT. Anti-NIP intensely stained the retina (F, R), optic nerve (G, ON), ciliary body (CB) and the lens (H, L), whereas sclera was negative.

The authors thank Svein A. Carlsen, Cancer Research Unit, Health Research Division, Saskatchewan Cancer Agency, for reading and critically evaluating the manuscript, and Todd Reichert for photography.