Ординатура / Офтальмология / Английские материалы / Progress in Lens and Cataract Research_Hockwin_2002

Our previous papers [4–6] have shown that in the human lens, ascorbate is maintained in the reduced state largely by ascorbate free radical (AFR) reductase (NADH:AFR oxidoreductase) in the soluble fraction, and its decline in activity is closely correlated with lens protein aggregation in age-related cataractogenesis and aging. We [7, 8] have further reported that major and minor AFR reductases are separated from the human lens soluble fraction by DEAEcellulose ion exchange column chromatography, and these AFR reductases also exhibit diaphorase activity using dichlorophenolindophenol and ferricyanide as electron acceptors.

The present investigation demonstrates that the soluble AFR reductases in the human lens are separated into many isoforms by isoelectric focusing, and that two main proteins with molecular weights of 20–25 kD are commonly identified to each isoform.

Materials and Methods

The major and minor AFR reductase fractions were isolated from the soluble fraction in the cortex of surgically enucleated human lenses with immature age-related cataract by DEAE-cellulose ion exchange column chromatography, as reported previously by us [7]. The isolated AFR reductase fractions were concentrated by ultrafiltration, dialyzed against 2 mM K phosphate, pH 7.2, and kept frozen at 80 °C until used.

Native isoelectric focusing (without 6 M urea) was carried out at a final voltage of 1,500 V for about 1.5 h at 10 °C using Ampholine PAGplate with a pH range of 3.5–9.5 (1-mm-thick polyacrylamide gel; Pharmacia Biotech, Uppsala, Sweden). After the focusing, the gel was stained for diaphorase activity with a zymogram technique or for a protein with Coomassie brilliant blue, as described previously [8], or used for the following twodimensional gel electrophoresis.

In the two-dimensional gel electrophoresis, the native isoelectric focusing was run as the first dimension. A 7-mm-wide strip was cut from the gel after the focusing. This strip was soaked for about 15 min in 10 ml of 0.125 M Tris-HCl, 2% SDS, pH 6.8 containing 0.25% dithiothreitol, and then for about 15 min in 10 ml of 0.125 M Tris-HCl, 2% SDS, pH 6.8 containing 4.5% iodoacetamide and 0.002% bromophenol blue. The treated gel strip was loaded onto ExcelGel SDS, gradient 8–18 (0.5-mm-thick polyacrylamide gradient gel for horizontal SDS electrophoresis; Pharmacia Biotech, Uppsala, Sweden), and the SDS gel electrophoresis for the second dimension was run at 600 V for about 2 h at 15 °C. After the electrophoresis, the gradient gel was stained for protein with Coomassie brilliant blue.

Results

Figures 1 and 2 show the results of native isoelectric focusing of the major and minor AFR reductase fractions isolated from the human lens soluble fraction by DEAE-cellulose ion exchange column chromatography. Many protein bands

pI 7.35

6.85

6.55

5.85

5.20

Fig. 1. Native isoelectric focusing of the major AFR reductase (7 g protein) isolated from the human lens soluble fraction. a Protein staining with Coomassie brilliant blue. b Diaphorase activity staining for1 h. c Diaphorase activity staining for 2 h.

are detected in the pI range of 5–7, and those from the major AFR reductase are more basic. Several protein bands in the pI range of about 6–7 from the major AFR reductase and in the pI range of about 5–6 from the minor AFR reductase are also stained for diaphorase activity. However, each of the isoforms with diaphorase activity appears to be somewhat different in the sensitivity of diaphorase activity.

Two-dimensional electrophoresis of the major and minor AFR reductase fractions, as shown in figures 3 and 4, reveals that two main protein spots with molecular weights of 20–25 kD are commonly identified to each of the isoforms. This result is consistent with that [8] in SDS polyacrylamide gel electrophoresis of a human lens AFR reductase (with a molecular weight of

pI

7.35

6.85

6.55

5.85

about 32 kD by gel filtration [7]) which was partially purified in 50-fold by three steps of column chromatography using DEAE-cellulose ion exchange, 5 AMP-Sepharose 4B affinity and Sephacryl S-200HR gel filtration columns.

Discussion

The present investigation manifests that AFR reductase in the human lens soluble fraction is very heterogeneous with regard to charge characteristics

pI

MW (kD)

94

67

43

30

20.1

14.4

Fig. 3. Two-dimensional gel electrophoresis of the major AFR reductase (33 g protein) isolated from the human lens soluble fraction. The gel was visualized for protein with Coomassie brilliant blue.

(fig. 1, 2). This heterogeneity may be associated with aging of the lens and/or cataractogenesis, since in this study, the enzyme has been isolated from the cortex of immature age-related cataractous human lenses.

However, the enzyme isoforms appear to be composed of similar protein subunits with molecular weights of 20–25 kD to each other (fig. 3, 4). Our previous paper [9] has demonstrated that AFR reductase (with a molecular weight of about 32 kD [7]) is eluted just after the peak of the dimeric -crystallin (with a molecular weight of about 40 kD [10]) in the gel filtration of the human lens soluble fraction on a Sephadex G-75 superfine column. 24and 27-kD protein subunits have been identified from the dimeric -crystallin [10]. These results suggest that the human lens AFR reductase may be a dimeric enzyme with a molecular weight of about 30–40 kD.

The observed heterogeneity of the human lens AFR reductase, which contains various more acidic components, is very similar to those of - and-crystallins in aged and cataractous human lenses as reported by Zigler et al. [10–12] and Kabasawa et al. [13]. Therefore, the lens AFR reductase as well as those crystallins may represent a long-lived protein, and some of the

pI

MW (kD)

94

67

43

30

20.1

14.4

Fig. 4. Two-dimensional gel electrophoresis of the minor AFR reductase (77 g protein) isolated from the human lens soluble fraction. The gel was visualized for protein with Coomassie brilliant blue.

isoforms, especially the more acidic isoforms of the minor AFR reductase (fig. 2), may be formed by posttranslational modifications associated with aging and cataractogenesis. We [8] have already reported that the specific activity of the minor AFR reductase is only about 20% of the major AFR reductase activity. Posttranslational modifications may cause a loss of enzyme activity, too. However, it is as yet unknown whether the more acidic isoforms of the minor AFR reductase increase with aging and/or cataractogenesis. Further investigation is under way to answer this question.

Acknowledgment

This investigation was supported in part by a joint research project from the Eye Research Institute of the Cataract Foundation, Tokyo, Japan.

References

1Reddy VN, Giblin FJ, Lin L-R, Chakrapani B: The effect of aqueous humor ascorbate on ultraviolet-B-induced DNA damage in lens epithelium. Invest Ophthalmol Vis Sci 1998;39:344–350.

2Nagaraj RH, Sell DR, Prabhakaram M, Ortwerth BJ, Monnier VM: High correlation between pentosidine protein crosslinks and pigmentation implicates ascorbate oxidation in human lens

senescence and cataractogenesis. Proc Natl Acad Sci USA 1991;88:10257–10261.

3Sasaki H, Giblin FJ, Winkler BS, Chakrapani B, Leverenz V, Chu-Chen S: A protective role for glutathione-dependent reduction of dehydroascorbic acid in lens epithelium. Invest Ophthalmol

Vis Sci 1995;36:1804–1817.

4Bando M, Obazawa H: Ascorbate free radical reductase and ascorbate redox cycle in the human lens. Jpn J Ophthalmol 1988;32:176–186.

5Bando M, Obazawa H: Activities of ascorbate free radical reductase and H2O2-dependent NADH oxidation in senile cataractous human lenses. Exp Eye Res 1990;50:779–784.

6Bando M, Obazawa H: Regional and subcellular distribution of ascorbate free radical reductase activity in the human lens. Tokai J Exp Clin Med 1991;16:217–222.

7Bando M, Obazawa H: Soluble ascorbate free radical reductase in the human lens. Jpn J Ophthalmol

1994;38:1–9.

8Bando M, Obazawa H: Ascorbate free radical reductases and diaphorases in soluble fractions of

the human lens. Tokai J Exp Clin Med 1995;20:215–222.

9Matsukura S, Bando M, Obazawa H: Ascorbate free radical reductase activities in soluble and

plasma membrane fractions of the lens. Atarashii Ganka 1999;16:383–386.

10Zigler JS Jr, Horwitz J, Kinoshita JH: Human -crystallin. I. Comparative studies on the 1, 2 and3-crystallins. Exp Eye Res 1980;31:41–55.

11Zigler JS Jr, Horwitz J, Kinoshita JH: Studies on the low molecular weight proteins of human lens. Exp Eye Res 1981;32:21–30.

12Zigler JS Jr, Russell P, Takemoto LJ, Schwab SJ, Hansen JS, Horwitz J, Kinoshita JH: Partial characterization of three distinct populations of human -crystallins. Invest Ophthalmol Vis Sci 1985;26:525–531.

13Kabasawa I, Kodama T, Kabasawa M, Sakaue E, Watanabe M, Kimura M: Heterogeneity of human cataractous and normal lens -crystallins. Exp Eye Res 1982;35:1–9.

Masayasu Bando, PhD, Department of Ophthalmology, Tokai University School of Medicine, Isehara, Kanagawa 259-1193 (Japan)

Tel. 81 463 93 1121, Fax 81 463 91 9328, E-Mail mbando@is.icc.u-tokai.ac.jp

and exhibits chaperone activity in vitro [4–6]. The first evidence of a chaperone role for -crystallin came from the observation by Horwitz that -crystallin could suppress the thermally induced aggregation of L- and -crystallin and that of a variety of enzymes [4]. Interestingly -crystallin protects - and-crystallin more efficiently than enzymes. Evidence for additional physiological roles for B-crystallin, a subunit of -crystallin, besides that of maintaining the structural integrity of the lens, has come from the finding that its expression, and that of A-crystallin, is not confined to the lens [7–9]. Elevated levels ofB-crystallin mRNA or protein are associated with a variety of pathological conditions [8, 10–13]. More recently, this protein has been shown to protect enzyme activity at physiological temperatures. It protected glucose-6-phosphate dehydrogenase [14] and malate dehydrogenase [15] against glycation-induced inactivation and catalase against steroid-induced inactivation [16]. -Crystallin also protected catalase against thermal inactivation [17].

The chaperone function of -crystallin is compromised with increasing age and in the diabetic lens. -Crystallin isolated from the lens of a newborn calf was a better chaperone than that isolated from an older bovine lens [18]. A decrease in the chaperone activity of -crystallin in older human [19] and diabetic rat [20] lenses has also been reported. Posttranslational modification of -crystallin with increasing age and in diabetes appears to be the reason for the decreased chaperone action but the main modification responsible for this change remains unresolved. The chaperone function of nucleus L-crystallin significantly declines with increasing age in both the rabbit [21] and human lens [22, 23]. No decline was seen in -crystallin isolated from cortical tissue.H-Crystallin had reduced chaperone function compared to L-crystallin in both cortex and nucleus but did not undergo a further significant decrease with age. With age the H-crystallin fraction and insoluble -crystallin increase while the soluble -crystallin fraction decreases [1, 24], which leads to an overall decline in chaperone activity.

-Crystallin would be expected to be susceptible to deleterious modification by metabolites in vivo with increasing age and in diseased conditions such as diabetes. However, -crystallin chaperone function has been shown to be resistant to several chemical modifications. Its chaperone function was not compromised when modified by galactose [25, 26]. No effect on chaperone activity was seen when -crystallin was modified by fructose, glucose 6-phosphate or by carbamylation [27], although glucose 6-phosphate and cyanate at higher concentrations can alter its secondary and tertiary structure [28, 29].

Recently Blakytny et al. [26] investigated the nature of modification of-crystallin by galactose. Likely sites for glycation appeared to be the lysine residues contained within the flexible tails of A- and especially B-crystallin.

Preferential glycation at the C-terminal extensions of -crystallin, a region thought to be important for chaperone activity, was found but the extensions maintained their flexibility and chaperone function was not compromised [26]. It may be that chaperone function is unimpaired as long as the C-terminal extensions remain flexible. In experiments with mutant -crystallins, introduction of a hydrophobic tryptophan in the tail region of A-crystallin greatly diminished chaperone activity and was associated with a loss of flexibility in the extension region [30].

In the results presented we show that modification of -crystallin by fructose 6-phosphate or prednisolone-21-hemisuccinate (p-21-h) compromised chaperone activity as judged by the L-crystallin thermal aggregation assay. However, in experiments using enzyme aggregation assays, chaperone activity was not compromised upon modification. Tryptophan fluorescence demonstrated that a conformational change may have occurred within the environment of a tryptophan residue of the modified -crystallin samples. This may indicate that the modification has taken place close to a tryptophan residue.

Materials and Methods

Materials

-Crystallin and L-crystallin were isolated from bovine lenses by gel chromatography on Sephacryl S300 HG using the method of Slingsby and Bateman [31]. Bovine liver catalase, equine liver alcohol dehydrogenase (ADH), bovine serum albumin, p-21-h, fructose 6-phosphate and all other proteins and chemicals were obtained from Sigma, Poole, Dorset, UK.

Incubation

Bovine -crystallin (4 mg/ml) was incubated at 37 °C with 25 mM fructose 6-phosphate for 30 days or with 10 or 25 mM p-21-h for 12 days in a shaking water bath. Solutions were dialysed over 48 h against four changes of 50 mM sodium phosphate buffer, pH 7, followed by four changes of distilled water, before being lyophilised.

Chaperone Assays

The effect of modification on the chaperone activity of bovine -crystallin was investigated using thermal aggregation assays. These assays followed the suppression by -crystallin of the thermally induced aggregation of L-crystallin [4] in 50 mM sodium phosphate buffer, pH 7, at 55 °C ( : L 1:12 w:w), ADH [4] in sodium phosphate buffer, pH 7, at 48 °C ( :ADH 1:3 w:w), glyceraldehyde 3-phosphate dehydrogenase ( :GAP-DH 1:2 w:w) in 100 mM triethanolamine buffer, pH 7.6 at 37 °C and catalase ( :CAT 1:12 w:w) in 50 mM sodium phosphate buffer, pH 7 at 55 °C. Triplicate experiments were carried out for each assay. Each figure representative of a thermal aggregation assay is the average of these three independent experiments.