Identification and Distribution of Dietary Precursors of The

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

Vision Research 39 (1999) 219–229

Identiﬁcation and distribution of dietary precursors of the Drosophila visual pigment chromophore: analysis of carotenoids in wild type and ninaD mutants by HPLC

David R. Giovannucci *, Robert S. Stephenson Department of Biological Sciences, Wayne State Uni6ersity, Detroit, MI 48202, USA

Received 3 July 1997; received in revised form 10 November 1997

Abstract

A dietary source of retinoid or carotenoid has been shown to be necessary for the biosynthesis of functional visual pigment in flies. In the present study, the larvae or adults of Drosophila melanogaster were administered specific carotenoid-containing diets and high performance liquid chromatography was used to identify and quantify the carotenoids in extracts of wild type and ninaD visual mutant flies. When i-carotene was fed to larvae, wild type flies were shown to hydroxylate this molecule and to accumulate zeaxanthin and a small amount of i-cryptoxanthin. Zeaxanthin content was found to increase throughout development and was a major carotenoid peak detected in the adult fly. Carotenoids were twice as effective at mediating zeaxanthin accumulation when provided to larvae versus adults. In the ninaD mutant, zeaxanthin content was shown to be specifically and significantly altered compared to wild type, and was ineffective at mediating visual pigment synthesis when provided to both larval and adult mutant flies. It is proposed that zeaxanthin is the larval storage form for subsequent visual pigment chromophore biosynthesis during pupation, that zeaxanthin or i-crytoxanthin is the immediate precursor for light-independent chromophore synthesis in the adult, and that the ninaD mutant is defective in this pathway. © 1998 Elsevier Science Ltd. All rights reserved.

1. Introduction dietary source for the generation of rhodopsin in flies (Zimmerman & Goldsmith, 1971; Harris, Ready, Lip- The identification of 11-cis 3-hydroxyretinal as the son, Huspeth & Stark, 1977; Stephenson, O’Tousa, visual pigment chromophore in Calliphora (Vogt, 1983; Scarvarda, Randall & Pak, 1983; Sapp, Christianson, Vogt & Kirschfeld, 1984) and in Drosophila (Gold- Maier, Studer & Stark, 1991). smith, Marks & Bernard, 1986; Seki, Fujishita, Ito, Since Vogt’s (1983, 1984) discovery that the fly’s Matsuoka & Tsukida, 1986; Tanimura, Isono & Tsuka- retinal chromophore is 3-hydroxylated, it has been pro- hara, 1986; Giovannucci & Stephenson, 1987), and the posed (and shown definitively in Calliphora) that flies elucidation of its recycling in the dipteran retina have the ability to convert h- and i-carotene to lutein (Schwemer, 1988) has aroused interest in the uptake, and zeaxanthin, respectively (Vogt & Kirschfeld, 1984), transport, and conversion of the carotenoid precursors and to use these polar derivatives to produce functional required for the biosynthesis of this novel chromophore visual pigment (Stark, Schilly, Christianson, Bone & and visual pigment. All higher animals studied to date, Landrum, 1990) or photoprotective pigment (Zhu & including flies, lack the ability to synthesize carotenoids Kirschfeld, 1984). or retinoids de novo and, thus, must obtain them To generate 11-cis 3-hydroxyretinal directly from i- through their diet (Kayser, 1982). An extensive body of carotene, a fly must perform at least three important i literature has established that -carotene can serve as a structural modifications on the carotene molecule: (1) hydroxylation at the 3 or 3% position of the ionone ring; (2) isomerization of the hydrocarbon chain at the 11 or * Corresponding author. Present address: Department of Physiol- 11% position; and (3) cleavage to a C20 molecule. How- ogy, University of Michigan Medical School, Ann Arbor, MI 48109, USA. Tel.: +1 313 7643339; fax: +1 313 9368813; e-mail: lud- ever, the exact sequence of these manipulations is not [email protected]. known. Zeaxanthin and lutein have been shown to be

0042-6989/98/$ - see front matter © 1998 Elsevier Science Ltd. All rights reserved. PII: S0042-6989(98)00184-9 220 D.R. Gio6annucci, R.S. Stephenson / Vision Research 39 (1999) 219–229 sufficient for the formation of functional rhodopsin in Table 1 Drosophila (Stark, Schilly, Christianson, Bone & Lan- Fly growth medium composition drum, 1990). Therefore, these hydroxylated, principally Sucrose-agar medium Cornmeal-agar medium all-trans carotenoids may function as substrates for the (g/lH2O) (g/lH2O) synthesis of the chromophore via a light-independent pathway, as demonstrated by the accumulation of 11- Agar19.8 68 cis 3-hydroxretinal in the heads of Drosophila reared on Yeast 126.4 193 xanthophylls (oxidized C40 compounds including hy- Sugar54 376 Proprionic acid11 ml 40 ml droxylated carotenoids) and maintained in the dark Yellow corn-0 773 (Seki, Fujishita, Ito, Matsuoka & Tsukida, 1986; Isono, meal Tanimura, Oda & Tsukahara, 1988). This would provide an alternative pathway to the adult fly’s blue light-dependent isomerization of all-trans retinal to the 2. Materials and methods 11-cis form (Schwemer, 1984; Isono, Tanimura, Oda & Tsukahara, 1988). However, the identification, quantifi- 2.1. Flies cation, and distribution of specific carotenoid subtypes (the presumed dietary sources for this unusual chro- Wild type and ninaD mutants (P246 and TH382 mophore) in Drosophila remain largely uncharacterized. The use of mutants that are defective in chromophore alleles) of Drosophila melanogaster were of the Oregon- biosynthesis will likely aid in determining the physio- R strain. The mutants were provided by Dr William logically important dietary substrates. Pak and Dr Theodore Homyk, Jr. of Purdue Univer- The Drosophila visual mutants designated ninaA sity. Both ‘wild type’ and mutant flies carried the through ninaH are characterized by reduced rhodopsin mutation (w) and therefore lacked screening pigment in levels and defective electroretinograms, indicating that the eye. a reduction in the amount of photopigment can occur through a variety of pathways (Pak, 1995). For exam- 2.2. Carotenoid-enriched and carotenoid-free diets ple, the 11-cis isomer of retinal or 3-hydroxyretinal has been shown to be necessary for the expression of func- The flies were reared on cornmeal-agar medium (the tional rhodopsin in flies (Schwemer, 1984; Isono, Tan- standard fly growth medium) or sucrose-agar medium imura, Oda & Tsukahara, 1988), and opsins with (Sang, 1956) (see Table 1). The former contains zeaxan- specific amino acid deletions or substitutions that result thin (Table 2), whereas the latter was free of in defective 11-cis retinal binding show altered cellular carotenoids or retinoids (data not shown) and was in processing and protein trafficking (Doi, Molday & some cases supplemented with pure carotenoid (100– Khorana, 1990; Anukanth & Khorana, 1994; Ridge, 500 mg/ml). i-carotene (Fluka), xanthophyll extract Lee & Abdulaev, 1996). Recently, evidence for altered (ICN), zeaxanthin, i-cryptoxanthin (gifts from Hoff- opsin processing has been demonstrated in ninaD flies mann-LaRoche), apo-8 i-carotenal (Fluka), or can- (Colley, Baker, Stamnes & Zuker, 1991; Ozaki, Na- thaxanthin (Fluka) was dispersed in sucrose-agar gatani, Ozaki & Tokunaga, 1993). Thus, genetic lesions medium that was warmed to a liquid. In general, flies that interfere with the availability of this isomer are were raised on a 12/12 h light/dark protocol at 22– likely to have profound effects on rhodopsin levels and 24°C or, in some cases, in complete darkness. Under on the function of the fly photoreceptor. Previously, we these conditions, the dispersed i-carotene had a half- have shown that the ninaD visual mutant expresses low life greater than 4 days at 22°C as determined by levels of rhodopsin when grown on standard (cornmeal- HPLC. In earlier experiments, flies were raised on agar) fly medium, but could be rescued (restored to the yellow cornmeal extract-enriched or carrot juice-en- wild type phenotype) by feeding with high levels of riched sucrose medium which contained substantial i-carotene or all-trans retinoids (Stephenson, O’Tousa, amounts of xanthophylls, carotenoid isomers, and Scarvarda, Randall & Pak, 1983; Giovannucci & other unidentified carotenoids. For the direct demon- Stephenson, 1988). This suggested that the ninaD mutant was deficient in some aspect of retinoid or Table 2 carotenoid metabolism, although the precise defect was Carotenoid content of cornmeal-agar medium (n=3) not determined. In the present study, high performance liquid chromatography (HPLC) was used to identify pmol/mg dry weight and quantitate the content and the distribution of 9 potential dietary carotenoid precursors of the fly visual Zeaxanthin 229 18 Lutein 91911 pigment chromophore in extracts of both wild type i-Cryptoxanthin 5397 Drosophila and ninaD larvae and adult flies fed on i-Carotene 43925 specific diets. D.R. Gio6annucci, R.S. Stephenson / Vision Research 39 (1999) 219–229 221

Table 3 Fly heads and bodies were separated by vortexing with Key for fly dietary history notation liquid nitrogen and shaking over a c30 mesh sieve. symbol Tissue was dried by blotting on filter paper, weighed, glass homogenized, and extracted twice with hexane/ Genotype acetone 1:1. If the tissue was not immediately used, it Wild type wt was sealed and stored at −70°C. For the analysis of P246 NinaD P246 carotenoids in fly culture media, a portion of medium NinaDTH382 TH382 was ground with two volumes of anhydrous Na2SO4 Stage and extracted with methylene chloride. B\ Larva 2.5. Chromatography Pupa () Adult (imago) {} Carotenoids and xanthophylls were reverse-phase Diet Carotenoid-free sucrose-agar medium (SAM) — chromatographed by the method of Nells & DeLeen- Cornmeal-agar medium cm heer (1983). Extracts were dried (Savant Speed-Vac) Carrot juice+SAM cj and the residue dissolved in 20-100 ml of mobile phase. i-carotene+SAM i-car Samples were injected onto a Zorbax-ODS column i i -cryptoxanthin+SAM -cry (Dupont) and eluted with 20% methylene chloride in Zeaxanthin+SAM Zea acetonitrile (v/v) at a flow rate of 1 ml/min. Absorbance For example, wt: Bi-car\(){–}, means a wild-type fly raised from was monitored at 450 nm. For the analysis of xantho- egg to larva on sucrose-agar medium containing, i-carotene then phylls only, we used a method similar to that described transferred prior to eclosion to sucrose-agar medium with no added by Almela, Lopez-Roca, Candela & Alcazar (1990). carotenoid. Samples were injected onto a Spherisorb Si column (phase separations) and eluted with hexane/acetone 4:1 i stration of the flies’ ability to structurally modify - (v/v) at 1 ml/min. After separation by HPLC, peaks m 14 i carotene, flies were raised on 100 g[ C] -carotene were collected and dried. Fractions were identified and m with a specific activity of 50 Ci/mg (gift from Dr Peter quantitated by their absorbance spectra (Hewlett-Pack- Sorter of Hoffmann-LaRoche) applied to the surface of ard Lambda 4 spectrophotometer) in hexane and by a sucrose-agar vial. comparison to known standards. Peak areas for known compounds were converted to picomoles from standard 2.3. Dietary notation curves for i-carotene, i-cryptoxanthin, or zeaxanthin using a molar extinction coefficient at 450 nm of Because various diets were administered at different 139000 l (mol/cm) (Floppen, 1971; Vetter, Englert, developmental time points, we have adopted a notation Rigassi & Schweiter, 1971). In some cases the peaks clearly identifying the specific dietary history of the were collected and identified by mass spectral analysis flies. This notation is explained in Table 3. (Enzell & Francis, 1969). Radio-chromatography was performed by mixing column effluent with an equal 2.4. Tissue preparation and extraction volume of scintillant (Flo-Scint I) in a Flo-One A265 radio-chromatographic detector (Radiomatic). Effi- When larvae were reared on high levels of i- ciency was determined to be 84%. All solvents for carotene, their guts appeared yellowish and contained extraction and for chromatography were of HPLC numerous orange-coloured lipid droplets. For this rea- grade and were degassed prior to use. The liquid chro- son, larvae raised on carotenoid-enriched media were matograph system consisted of an LKB model 2150 transferred to sucrose-agar medium prior to collection. pump, 2152 pump controller, and 2151 variable wave- In most experiments, the third instar larvae were re- length detector or a radiometric detector. Peaks were moved from the supplemented medium and placed on integrated with chromatography software programs sucrose-agar medium. After 24 h on this medium, the (Nelson Analytical or Radiomatic). larval gut appeared by visual inspection to be clear of For the analysis of retinoids by HPLC, wild type and pigmented material. Moreover, a comparison by HPLC ninaD mutant fly heads were prepared as described of several unidentified non-carotenoid peaks in the above. Tissue extraction, preparation of retinal oximes, extracts of flies, maintained on cornmeal medium with and chromatography of 3-hydroxyretinoids was the flies that were tranferred to sucrose medium, revealed same as that described by Groenendijk, DeGrip & that \95% of the material in the gut had been ex- Daemen (1979) and Goldsmith, Marks & Bernard changed. For the HPLC analysis of carotenoids, pupae (1986). Chromatograms were developed isocratically were collected with cotton swabs just prior to eclosion. with 15% dioxane in n-hexane at a flow rate of 1 The adults were collected within 1 day after eclosion or ml/min on a Spherisorb 5 mm silica column (phase maintained on sucrose-agar medium for several days. separations) and the eluent monitored at 338 nm. 222 D.R. Gio6annucci, R.S. Stephenson / Vision Research 39 (1999) 219–229

2.6. Statistics these retinoids. Conversely, hydroxyretinoids were never detected outside the head in either wild type, All of the data are expressed as the mean and stan- mutant flies, or in fly larvae. dard error, where n represents the number of experiments. For example, for the HPLC analysis of retinoid 3.2. Carotenoids in 6arious media content, each experimental batch usually contained between 100–300 fly heads. Baseline separation could be achieved for most carotenoids of interest. Elution times ranged from 3 to 29 min for both non-aqueous reverse phase and normal 3. Results phase systems. Fig. 2 (top panel) shows that baseline resolution of i-carotene, i-cryptoxanthin, and zeaxan- 3.1. 3-Hydroxyretinoids in mutant flies thin standards in mixture could be achieved using the normal phase chromatography system. However, can- When raised on the standard fly medium, ninaD thaxanthin and i-cryptoxanthin co-eluted. All of these mutants are deficient in 11-cis 3-hydroxyretinal (Gio- compounds were resolvable under reverse phase chro- vannucci & Stephenson, 1987). Consistent with the requirement of 11-cis 3-hydroxyretinal for the biosynthesis and trafficking of functional visual pigment in flies, Colley, Baker, Stamnes & Zuker (1991) and Ozaki, Nagatani, Ozaki & Tokunaga (1993) have shown that ninaD flies express a 40 kD, glycosylated, immature form of rhodopsin. However, it was not clear just where the ninaD defect exerted its effect. Also, it was not known if other all-trans and 11-cis hydroxyretinoids, which have been reported to occur in Drosophila (Goldsmith, Marks & Bernard, 1986; Seki, Fujishita, Ito, Matsuoka, Kobayshi & Tsukida, 1987), represented precursors or byproducts of chromophore biosynthesis. To clarify the ninaD defect, we investigated the hydroxyretinoid content of wild type and mutant flies by HPLC. It was found that wild type (wt): Bcm\(){cm} heads contained on average 1.2890.04 pmol of 3-hydroxyretinal (n=13) consistent with previous reports (Goldsmith, Marks & Bernard, 1986; Seki, Fujishita, Ito, Matsuoka & Tsukida, 1986; Tanimura, Isono & Tsukahara, 1986; Giovannucci & Stephenson, 1987). By contrast, wt: B – \(){–} heads exhibited less than 1% of their normal complement of 11-cis 3-hydroxyretinal. However, when these deprived adult flies were placed on medium enriched with high levels of i-carotene, their 3-hydroxyretinoid levels returned to near normal within several days (0.98190.08 pmol; Fig. 1. Chromatograms of (a) wild type Drosophila heads from flies n=5). Fig. 1(a,c,d) show representative chro- P246 TH382 P246 reared on standard cornmeal-agar medium; (b) ninaD heads from matograms of wild type, ninaD , and ninaD flies rescued by raising adults on i-carotene enriched medium; (c) Bcm\(){–} flies. In the mutants, the peaks corre- ninaDTH382 heads; and (d) ninaDP246 heads from flies reared on sponding to 3-hydroxyretinoid were largely absent (at- standard medium (not rescued). Homogenates of whole heads were tenuated 100-fold compared to the wild type fly). There treated with hydroxylamine and extracted for HPLC analysis. Peaks was only about 0.02 pmol of 3-hydroxyretinal in each are identified as (1) syn 11-cis 3-hydroxyretinal oxime; (2) syn all- B trans 3-hydroxyretinal oxime; (3) anti all-trans 3-hydroxyretinal mutant head. However, Fig. 1b shows that P246: – oxime; and (6) anti 11-cis 3-hydroxyretinal oxime. Peaks (4) and (5) \(){i-car} had hydroxyretinoid profiles that did not fluoresce and are likely isomers of 3-hydroxyretinol. It should be differ significantly from wild type flies. These flies con- noted that the chromatograms of fly extracts were run on different tained about 1.190.04 pmol 3-hydroxyretinal per head days. This likely accounts for the differences in peak retention times (n=9). It is important to note that non-hydroxylated between (a) and (b). However, the major hydroxyretinoid peak in both was identified as all-trans 3-hydroxyretinal oxime, based on the retinoids were never detected in extracts from any of retention time of a 3-hydroxyretinal oxime standard. Column, these heads, even when the extraction and chromatog- Spherisorb Si; mobile phase, 84% hexane, 16% dioxane (v/v); detec- raphy conditions were optimized for the detection of tion, 338 nm; flow rate, 1 ml/min. D.R. Gio6annucci, R.S. Stephenson / Vision Research 39 (1999) 219–229 223

Fig. 2. Chromatograms of mixtures of carotenoid standards separated by normal and reverse phase HPLC. Peaks are identified in both chromatograms as canthaxanthin (CAN), zeaxanthin (ZEA), i-cryptoxanthin (i-cry), and i-carotene (i-car). Inset at right shows the chemical structures of i-carotene and the important xanthophylls. Normal phase column: Spherisorb-Si column, hexane/acetone 4:1 (v/v);1ml/min flow. Reverse phase column: Zorbax-ODS, 20% methylene chloride in acetonitrile (v/v);1ml/min flow, detection at 450 nm. Time scale is the same for both figures. matography (Fig. 2, bottom panel), although the run and fungi can synthesize carotenoids de novo, bacterial time for chromatography was nearly doubled. The nor- cultures were isolated from flies and stock bottles and mal phase system was our usual choice for analysis. In investigated as a possible source of carotenoids. Cul- some experiments, however, both systems were used to tures grown on blood-agar or in broth were saponified quantify the carotenoids present in the various diets with ethanolic KOH, extracted with hexane, and ana- presented to flies. Carotenoid isomers were not resolv- lyzed by HPLC. No measurable amounts of able on either system. The standard medium on which carotenoids were detected from any culture. As such, it flies were reared was cornmeal-agar medium and con- was not necessary to add antibiotics or fungal in- tained mostly xanthophylls (Table 2). Since bacteria hibitors to the fly medium. 224 D.R. Gio6annucci, R.S. Stephenson / Vision Research 39 (1999) 219–229

3.3. Carotenoids in lar6ae pupa presumably far exceeds the amount necessary for the synthesis of wild type levels of visual pigment In our initial experiments, flies were raised on and chromophore, if one assumes that two molecules of collected from carrot juice-enriched medium (Bcj\ hydroxyretinal can be produced from one molecule of (){cj}). Drosophila larvae were found to absorb large i-carotene. It is not clear why larvae store such large amounts of dietary carotenoids such as i-carotene, amounts of i-carotene or purge themselves prior to h-carotene, zeaxanthin, lutein, i-cryptoxanthin, i-apo pupation. 8-carotenal, canthaxanthin, and carotenoid epoxides with relative indiscrimination. Fig. 3 shows that both 3.4. Carotenoids in adult flies wt: Bi-car\ and P246:Bi-car\ third instars sequestered about 1 nmol carotenoid/larva. i-carotene A newly formed wild type fly emerges from its pupar- comprised 93% of the total carotenoid in wild type and ium with a full measure of visual pigment. Since flies 98% in ninaD P246. The remaining carotenoid in this set cannot synthesize carotenoids or retinoids de novo, larvae raised on carotenoid-containing medium must of experiments was not characterized, although it likely carry a carotenoid, or some carotenoid derivative, with consisted of xanthophylls and epoxides. The upper limit them into pupation. It is from this pool that the of total carotenoid sequestered by larvae starved prior chromophore is synthesized, perhaps in concert with to analysis is from 1- to 4-fold the concentration of the development of the adult eye from the eye imaginal carotene in the medium. disc. Although it has been reported that there is a general Like larvae, wt: Bcj\() and P246:Bcj\() pupae increase in the lipid content, water loss and lipid hy- contained mostly i-carotene. The amount of i- drolysis contributes to a loss in weight of the pre-pupa. carotene did not differ significantly between wild type Additionally, there is about a 30% reduction in body and mutant (44920 pmols/pupa (n=12), and 53936 weight between the pupa and the adult fly, of which pmols/pupa (n=8), respectively). At eclosion, however, 80% is associated with the puparium (Church & more than 70% of i-carotene was associated with the Robertson, 1966). Most of the carotenoid sequestered puparium, and the pool of carotenoid remaining with by larvae was excreted just prior to pupation. Even so, the wild type adult was enriched in the hydroxylated the remaining carotenoid retained by the wild type forms. Therefore, we used HPLC to identify and investigate the distribution of xanthophylls in both wild type and mutant adult flies. Wild type and ninaD mutant adult flies were shown to contain both i-carotene and xanthophylls (Fig. 3). However, on average, wild type and mutant adult flies differed in the ratio of i-carotene to xanthophylls, and in the total amount of xanthophylls extractable. Wt: Bcj\(){cj} and P246:Bcj\(){cj} whole flies contained about 62 pmol (n=11) or 40 pmol (n=6) of total carotenoid, respectively. i-carotene comprised 30% of the total carotenoid in wild type flies and nearly 60% of the total in the ninaD P246 mutant. The remaining carotenoid in these samples consisted of a number of polar compounds, some of which were identified on the basis of retention time as lutein, zeaxanthin, and i-cryptoxanthin. Wild type decapitated bodies contained 42912 pmol (n=11) and wild type heads contained 1.990.5 pmol (n=4) of xanthophyll, wheareas, ninaD P246 bodies and heads contained only 1796 pmol (n=6) and 0.490.4 pmol (n=2) of xanthophyll, respectively. (Lutein and h-carotene were detectable only in flies that were raised on carrot juice or cornmeal- agar medium.) Thus, compared to wild type flies, both Fig. 3. Total carotenoid content of wild type (wt) and ninaDP246 (nD) the decapitated bodies and the heads of the ninaD P246 Drosophila and at different developmental stages. Fruitflies were flies showed a reduction in xanthophyll content. This maintained on carrot juice-enriched sucrose-agar medium (Bcj\ (){cj}). Unfilled bars represent carotene content (i- and h-carotene) reduction in xanthophylls was also evident in P246 and hatched bars represent xanthophyll content (zeaxanthin, lutein, ninaD :Bi-car\(){–} and Bcm\(){–} flies (see and i-cryptoxanthin). below). D.R. Gio6annucci, R.S. Stephenson / Vision Research 39 (1999) 219–229 225

Fig. 4. Representative chromatograms of carotenoids extracted from fly head homogenates. Flies were reared on cornmeal medium and transferred to sucrose medium (to clear their guts of residual carotenoid) for 1 day prior to extraction. The peak corresponding to zeaxanthin is greatly enhanced in the ninaDTH382 mutant and greatly attenuated in the ninaDP246 mutant compared to the wild type. Chromatograms of (a) 425 ninaDTH382 heads; (b) 1250 ninaDP246 heads; and (c) 800 wild type Drosophila heads. Peaks are identified as (1) i-cryptoxanthin; (2) zeaxanthin; and (3) lutein. Column, Spherisorb Si; eluent, 20% acetone in hexane; flow rate, 1 ml/min; detection, 450nm.

Because xanthophylls are the major carotenoid treated ninaD P246 flies contained only 0.03 pmol of source found in the standard medium on which flies are zeaxanthin/fly (n=6). In comparison, ninaD TH382 flies raised, the consequence of a defect in the ability to use contained significantly more zeaxanthin than wt: B xanthophylls would be to render the flies vitamin A- cm\(){–}: 10.590.7 pmol/whole fly (n=5). Further- deficient. We hypothesized that the ninaD defect inter- more, the ninaD TH382 head contained 2.790.03 pmol fered with the flies’ ability to use xanthophylls, and of zeaxanthin, or an amount nearly ten times that tested the xanthophyll content of two different alleles of extractable from a wild type head. The concentration of this mutation: ninaD P246 and ninaD TH382. The two alle- zeaxanthin (in pmol/mg dry weight) in ninaD TH382 fly les showed altered xanthophyll content, either a significant reduction or enhancement in the amount as compared to wild type. An altered carotenoid profile was most evident in flies that were grown as larvae on cornmeal-agar medium. For example, Fig. 4 shows representative chromatograms identifying the carotenoids present in TH382, P246, or wt: Bcm\ (){–} flies (extracts of 425, 1250, and 800 heads, respectively). The heads of ninaD P246 flies exhibited approximately a 100-fold reduction in the amount of hydroxylated carotenoid. Conversely, the ninaD TH382 showed a 10-fold increase compared to wild type levels. This data suggested that the effect of the ninaD mutation is to interfere with either the filling or use of an available pool of zeaxanthin, or a derivative of zeaxanthin, that serves as an important precursor for the production of visual pigment chromophore. Fig. 5 Fig. 5. Xanthophyll content of wild type ninaDP246, and ninaDTH382 B \ flies reared as larvae on standard cornmeal-agar medium or on pure shows that, on average, wt: cm (){–} Drosophila i 9 -carotene-enriched sucrose-agar medium (200–300 mg carotenoid/ contained 3.4 0.5 pmol/fly of zeaxanthin (with a trace ml medium). Zeaxanthin was the major carotenoid present in wild amount of i-cryptoxanthin) (n=6), wheareas, similarly type or TH382 flies reared on cornmeal-agar medium. 226 D.R. Gio6annucci, R.S. Stephenson / Vision Research 39 (1999) 219–229 heads was 6-fold greater than that found in the rest of the body. The high levels of zeaxanthin in ninaD TH382 is consistent with the observation that this fly’s eye has a yellowish cast when compared to wild type. This head- specific enrichment of zeaxanthin, however, was not observed for the wild type or for the ninaD P246 flies. (Although quantitation of the reduced content of P246: Bcm\(){–} makes a comparison difficult to assess.) The data presented above demonstrated that wild type larvae could absorb and store sufficient amounts of i-carotene and zeaxanthin to supply the adult fly with the necessary precursor for chromophore synthesis. However, the relative efficacy of these two compounds when fed to ninaD mutant larvae was not known. Therefore, wild type and ninaD larvae were reared on i-carotene-enriched diets to investigate the effectiveness of i-carotene to serve as a substrate for zeaxanthin biosynthesis. After eclosion, HPLC was used to determine the amount of xanthophyll in Bi- car\(){–} flies and compared to Bcm\(){–} flies. In Bi-car\(){–} flies, i-carotene was the major carotenoid present and the remaining carotenoid (18%) was mostly zeaxanthin, with only trace amounts of i-cryptoxanthin detected. The total amounts of xanthophylls extracted from wild type, ninaD P246, and ninaD TH382 flies were 1.890.1, 0.490.4, and 4.290.2 pmol (8]n]2), respectively. Fig. 5 compares the amounts of xanthophylls (mostly zeaxanthin) extracted from Bcm\(){–} and Bi-car\(){–} flies. These experiments demonstrated that Drosophila could hydroxylate i-carotene. Furthermore, the ninaD: Bi- car\(){–} flies showed either a reduction (P246) or an enrichment (TH382) in zeaxanthin content. However, when reared on i-carotene, the pool of zeaxanthin was only about half that of Bcm\(){–} mutants. Finally, the ability of Drosophila to produce significant amounts of zeaxanthin from i-carotene was directly demonstrated in an experiment where wild type Fig. 6. Radiochromatograms of (a) carotenoids from wt: B – \(){–} Drosophila were fed [14C] i-carotene (Fig. 6). In this fed [14C] i-carotene (100 mg/ml) for 1 week; and (b) [14C] i-carotene experiment wt: B – \(){–} flies were fed radioactive standard. C, absorbance detection trace corresponding to radiochro- i-carotene for 4 days and then transferred to matogram (a). Peaks were identified as carotenoids by their spectra and retention time. Peak 1, i-carotene; Peak 2, i-cryptoxanthin; carotenoid-free medium for 1 day before analysis by Peak 3, zeaxanthin; Peak ?, was not identified but was likely a radio-chromatography. Whole fly extracts showed that breakdown product of [14C] i-carotene. Column, spherisorb-Si; see significant amounts of radiolabeled zeaxanthin (40% of text for conditions. the total extractable carotenoid) and a small amount of i-cryptoxanthin, in addition to i-carotene, were accu- This work indicates that retinoids are not likely to be mulated after a few days. the major dietary source for fly chromophore production. For example, wt: Bi-car\(){–} flies were found 4. Discussion to have 3-hydroxyretinoids but no detectable levels of retinoids or retinyl esters (Giovannucci & Stephenson, Using HPLC to analyze extracts of flies raised on 1987). In addition, 3-hydroxyretinoid was not found in specific diets, the present study provides information on the larva or in the adult fly outside the head. Previously the occurrence, content, and utilization of carotenoids we had shown that when provided to wild type larvae, during the Drosophila life cycle. In addition, we present all-trans retinoids were poor mediators of visual pig- evidence that the Drosophila ninaD visual mutant is ment biosynthesis and flies fed these compounds defective in xanthophyll metabolism. emerged from their puparia with low visual pigment D.R. Gio6annucci, R.S. Stephenson / Vision Research 39 (1999) 219–229 227 levels (Giovannucci & Stephenson, 1988). On the other carotenoid, zeaxanthin, is an important physiologically hand, wild type flies raised on the standard medium active molecule for chromophore biosynthesis. For ex- (which contained no appreciable amount of vitamin A) ample, zeaxanthin was shown to be the major emerged from their puparia with a full complement of carotenoid peak detected in extracts of Bcm\(){–} functional visual pigment. However, in the puparium, flies and was also the most altered in the ninaD visual the light required to effectively drive the isomerization mutant. Zeaxanthin was a major peak even when flies of all-trans hydroxyretinal to the 11-cis form (Schwe- were raised on i-carotene-enriched medium (with no mer, 1984; Isono, Tanimura, Oda & Tsukahara, 1988) zeaxanthin). On this diet, the relative proportion of would presumably be attenuated. This suggests that zeaxanthin to i-carotene was shown to be enriched there exists another system for chromophore biosynthe- about 4-fold during development (between larvae and sis that does not rely on light-evoked isomerization. flies). We do not mean to imply that zeaxanthin is Adult flies have been shown to have a light-dependent necessarily the direct precursor of 11-cis 3-hydroxyreti- system for chromophore regeneration. Adult nal. Rather, it is more likely that zeaxanthin is a Drosophila, maintained on carotenoid in the dark accu- preferred storage form. There are other possible precur- mulate 11-cis 3-hydroxyretinal (Seki, Fujishita, Ito, sors. The monohydroxylated carotenoid i-cryptoxan- Matsuoka & Tsukida, 1986; Isono, Tanimura, Oda & thin was also present in extracts of flies, although at Tsukahara, 1988), indicating that carotenoids may much lower levels than zeaxanthin. i-cryptoxanthin provide a direct, light-independent pathway for chro- content was also reduced in the P246 mutant, and mophore biosynthesis to bypass a light-driven isomer- interestingly, in the head but not the body of the ization. Furthermore, a light-independent isomerase TH382 mutant, and application of i-cryptoxanthin to would not necessarily have to be located in the eye or in the fly cornea supports robust visual pigment biosyn- the head. In the current study, however, we were not thesis (data not shown). Furthermore, the low levels of able to resolve carotenoids isomers using HPLC, and i-cryptoxanthin in xanthophyll-fed flies could indicate spectrophotometric data were inconclusive as to the that the removal of a hydroxyl group from zeaxanthin isomeric configuration of purified carotenoids. As such, is a regulated metabolic step or that i-cryptoxanthin, we were unable to demonstrate the presence of cis once produced, is rapidly used. carotenoids. Therefore, the location of a light-indepen- The majority of visual pigment biosynthesis in the fly dent isomerase remains to be determined. By contrast, occurs during late pupation when the adult eye is the cleavage of the hydroxylated C40 molecule likely formed from the eye imaginal disc. Since this occurs at occurs in the head because retinoids were never de- a time when the fly is cloistered in its puparium, the tected outside the fly head. larval fly must store, and the pupa subsequently liber- In the present study we attempted to determine ate, the compounds necessary to support visual pigment which carotenoid was the physiologically significant synthesis. Perhaps it is for this reason that stores of form. Although i-carotene has been shown to mediate hydroxylated carotenoids were significantly greater 3-hydroxyretinal synthesis in flies, the present study when flies were fed carotenoids while larvae rather than indicates that i-carotene itself is not enzymatically adults. Little is known about the fly’s ability to store, cleaved and is not likely to be a direct carotenoid liberate, and transport these highly hydrophobic com- substrate for chromophore generation. This is sup- pounds to their appropriate sites of action. We found ported by the absence of standard (non-hydroxylated) no evidence of carotenyl esters, nor was there any retinoids (the expected products of a i-carotene dioxy- obvious localization of carotenoids. Except in the case genase) (Goodman et al., 1966) and by the observation of ninaD TH382, the concentrations of i-carotene, i- that i-carotene was never detected in flies that were cryptoxanthin, and zeaxanthin were uniform between administered xanthophyll diets. Many dipterans, how- the head and body. However, the demonstration of ever, possess the ability to introduce hydroxyl groups to lipid/carotenoid transport proteins in the banana fly the ionone ring at the C3 position (Kayser, 1985; Smith Rhynchosciara americana (Terra, De Bianchi, Paes de & Goldsmith, 1990), and this was conclusively shown Mello & Basile, 1976) and the recent identification of a for the fly Calliphora (Vogt & Kirschfeld, 1984). Simi- lutein-binding protein in the midgut of the silkworm larly, we found a direct demonstration of the ability of (Jouni & Wells, 1996), and a Drosophila homolog of the Drosophila to modify dietary i-carotene by hydroxyla- interphotoreceptor retinoid binding protein (DRBP) tion (Fig. 6). We found that Drosophila will absorb (Kutty, Kutty, Duncan, Rodriguez, Shim, Stark & Wig- carotenoids with relative indiscrimination, and previous gert, 1996) suggests that their may exist specific binding reports have demonstrated that i-carotene, zeaxanthin, proteins for the transport of carotenoids form specific and lutein are equally effective at promoting visual cellular or subcellular compartments, or through the pigment synthesis in carotenoid-deprived adult flies hemolymph. We suggest that the altered zeaxanthin (Stark, Schilly, Christianson, Bone & Landrum, 1990). content of the ninaD mutants may result from a defec- The present data indicate that the dihydroxylated tive xanthophyll binding or transport protein. This 228 D.R. Gio6annucci, R.S. Stephenson / Vision Research 39 (1999) 219–229 could account for the opposite effects on zeaxanthin Colley, N. J., Baker, E. K., Stamnes, M. A., & Zuker, C. S. (1991). content observed for the two alleles. A binding protein The cyclophylin homolog ninaA is required in the secretory pathway. Cell, 67, 255–263. that binds too weakly or too strongly (that sequesters Doi, T., Molday, R. S., & Khorana, H. G. (1990). Role of the or misdirects large amounts of its ligand) would, in the intradiscal domain in rhodopsin assembly and function. Proceed- final analysis, deprive the animal equally. ings of the National Academy of Sciences USA, 87, 4991–4995. Because the ninaD mutant resembles a vitamin A-de- Enzell, C. R., & Francis, G. W. (1969). Mass spectrometric studies of prived wild type fly, and can be rescued by appropriate carotenoids 2. Survey of fragmentation reactions. Acta Chemica Scandina6ica, 23, 727–750. dietary supplementation, it has been referred to in the Floppen, F. H. (1971). Tables for the identification of carotenoid literature as a retinoid mutant. In fact, both the pigments. Chromatography Re6iews, 14, 133–298. ninaD P246 and ninaD TH382 can handle retinoids at least Giovannucci, D. R., & Stephenson, R. S. (1987). Analysis by HPLC as well as similarly treated vitamin A-deprived wild of Drosophila mutations affecting vitamin A metabolism. In6es- tigati6e Ophthalmology and Visual Science (Suppl., 28, 253. type flies. Furthermore, the retinoid content of heads of Giovannucci, D. R., & Stephenson, R. S. (1988). The Drosophila the mutant flies raised as adults on retinoid-enriched visual mutation affects vitamin A uptake. In6estigati6e Ophthal- medium does not significantly differ from that of the mology and Visual Science (Suppl.), 29, 388. wild type. However, in light of the differences in the Goldsmith, T. H., Marks, B. C., & Bernard, G. D. (1986). Separation hydroxylated carotenoid content observed between ni- and identification of geometric isomers of 3-hydroxyretinoids in the eyes of insects. Vision Research, 26, 1763–1769. naD and wild type flies, and between the two alleles, it Goodman, de W. S., Huang, H. S., & Shiratori, T. (1966). Mecha- is suggested that the ninaD mutation be reconsidered as nism of biosynthesis of vitamin A from i-carotene. Journal of a mutation that affects carotenoid metabolism. The Biological Chemistry, 241, 1929–1932. ninaD lesion affects the storage and utilization of xan- Groenendijk, G. W., DeGrip, W. J., & Daemen, F. J. (1979). thophylls such as i-cryptoxanthin and, in particular, Identification and characterization of syn- and anti-isomers of retinal oximes. Analytical Biochemistry, 99, 304–310. zeaxanthin. The defect exerts its effect prior to eclosion Harris, W. A., Ready, D. F., Lipson, E. D., Huspeth, A. J., & Stark, and interferes with (adult effect) or blocks (larval effect) W. S. (1977). Vitamin A derivation and Drosophila photopig- the supply of a component necessary for the synthesis ments. Nature, 266, 648–650. of visual pigment. Since zeaxanthin is the only physio- Isono, K., Tanimura, T., Oda, Y., & Tsukahara, Y. (1988). Depen- dency on light and vitamin A derivatives on the biogenesis of logically active carotenoid that was not effective at 3-hydroxyretinal and visual pigment in the compound eyes of mediating visual pigment synthesis when provided to Drosophila melanogaster. Journal of General Physiology, 92, 587– both ninaD larvae and adult, we suggest that zeaxan- 600. thin is the larval storage form and that zeaxanthin or Jouni, Z. E., & Wells, M. A. (1996). Purification and partial charac- i-cryptoxanthin is the substrate for a light-independent terization of a lutein-binding protein from the midgut of the silkworm Bombyx mori. Journal of Biological Chemistry, 271, isomerase and 11-cis 3-hydroxyretinal generation. 14722–14726. Kayser, H. (1982). Carotenoids in insects. In G. Britton, & T. W. Goodwin, Carotenoid chemistry and biochemistry. New York: Acknowledgements Pergamon, 195–210. Kayser, H. (1985). Pigments. In G. A. Kerkut, & L. I. Gilbert, 6 The authors would like to thank Dr William Pak and Comprehensi e insect physiology, biochemistry, and pharmacology, Vol. 10. New York: Pergamon, 367–415. Dr Theodore Homyk, Jr. for providing the ninaD mu- Kutty, R. K., Kutty, G., Duncan, T., Rodriguez, I. R., Shim, K., tants P246 and TH382, Dr Richard Bone and Dr John Stark, W. S., & Wiggert, B. (1996). Molecular cloning of a novel Landrum for providing pure xanthophyll standards, gene encoding a Drosophila retinoid- and fatty acid-binding glyco- Hoffmann LaRoche & Co, Nutley, NJ for the radiola- protein expressed in cone (semper) cells of the compound eye. i In6estigati6e Ophthalmology and Visual Science (Suppl.), 37, 802. beled -carotene, and Dr Stamatina Ziemba for the Nells, H. J. C. F., & DeLeenheer, A. P. (1983). Isocratic nonaqueous microbiological analysis of the fly medium. This work reversed-phase liquid chromatography of carotenoids. Analytical was supported by NIH grants EY-04186 and RR08167 Chemistry, 55, 270–275. to RSS. Ozaki, K., Nagatani, H., Ozaki, M., & Tokunaga, F. (1993). Matura- tion of major Drosophila rhodopsin, ninaE, requires chromophore 3-hydroxyretinal. Neuron, 10, 1113–1119. Pak, W. L. (1995). Drosophila in vision research. The Friedenwald References Lecture. In6estigati6e Ophthalmololgy and Visual Science, 36(12), 2340–2357. Almela, L., Lopez-Roca, J. M., Candela, M. E., & Alcazar, M. D. Ridge, K. D., Lee, S. S. J., & Abdulaev, N. G. (1996). Examining (1990). Separation and determination of individual carotenoids in rhodopsin’s folding and assembly through expression of polypep- a Capsicum cultivar by normal-phase high-performance liquid tide fragments. Journal of Biological Chemistry, 271, 7860–7867. chromatography. Journal of Chromatography, 502, 95–106. Sang, J. H. (1956). The quantitative nutritional requirements of Anukanth, A., & Khorana, H. G. (1994). Structure and function in Drosophila melanogaster. Journal of Experimental Biology, 33, rhodopsin. Requirements of a specific structure for the intradiscal 45–72. domain. Journal of Biological Chemistry, 269, 19738–19744. Sapp, R. J., Christianson, J. S., Maier, L., Studer, K., & Stark, W. S. Church, R. B., & Robertson, F. W. (1966). A biochemical study of (1991). Carotenoid replacement therapy in Drosophila: recovery the growth of Drosophila melanogaster. Journal of Experimental of membrane, opsin, and visual pigment. Experimental Eye Re- Zoology, 162, 337–352. search, 53, 73–79. D.R. Gio6annucci, R.S. Stephenson / Vision Research 39 (1999) 219–229 229

Schwemer, J. (1984). Renewal of visual pigment in photoreceptors of Tanimura, T., Isono, K., & Tsukahara, Y. (1986). 3-hydroxyretinal as the blowfly. Journal of Comparati6e Physiology A, 154, 535–547. a chromophore of Drosophila melanogaster visual pigment ana- Schwemer, J. (1984). Cycle of 3-hydroxyretinoids in an insect eye. In lyzed by high-pressure liquid chromatography. Photochemistry T. Hara, Molecular physiology of retinal proteins. Osaka: Yamada and Photobiology, 43, 225–228. Science Foundation, 299–304. Terra, W. R., De Bianchi, A. G., Paes de Mello, M., & Basile, R. Seki, T., Fujishita, S., Ito, M., Matsuoka, C., & Tsukida, K. (1986). (1976). Some metabolic disorders affecting the carotenoid-linked A fly, Drosophila melanogaster, forms 11-cis 3-hydroxyretinal in haemolymph proteins in Rhynchosciara americana (Diptera, Scia- the dark. Vision Research, 26, 255–258. ridae). Experientia, 32, 432–434. Seki, T., Fujishita, S., Ito, M., Matsuoka, N., Kobayshi, C., & Vetter, W., Englert, G., Rigassi, N., & Schweiter, U. (1971). Spectro- Tsukida, K. (1987). Retinoid composition in the compound eyes scopic methods. In O. Isler, Carotenoids. Basel: Birkhauser, 189– of insects. Experimental Biology, 47, 97–103. 266. Smith, W. C., & Goldsmith, T. H. (1990). Phyletic aspects of the Vogt, K. (1983). Is the fly visual pigment a rhodopsin? Zeitschrift fuer distribution of 3-hydroxyretinal in the class Insecta. Journal of Naturforschung, 38C, 329–333. Molecular E6olution, 30, 72–84. Vogt, K. (1984). The chromophore of the visual pigment in some Stark, W. S., Schilly, D., Christianson, J. S., Bone, R. A., & Lan- drum, J. T. (1990). Photoreceptor specific efficiencies of i- insect orders. Zeitschrift fuer Naturforschung, 39C, 196–197. carotene, zeaxanthin, and lutein for photopigment formation Vogt, K., & Kirschfeld, K. (1984). Chemical identity of the chro- deduced from receptor mutants of Drosophila melanogaster. Jour- mophore of fly visual pigment. Naturwissenschaften, 71, 211–213. nal of Comparati6e Physiology A, 166, 429–436. Zhu, H., & Kirschfeld, K. (1984). Protection against photodestruc- Stephenson, R. S., O’Tousa, J. E., Scarvarda, N. J., Randall, L. L., & tion in fly photoreceptors by carotenoid pigments. Journal of Pak, W. L. (1983). Drosophila mutants with reduced rhodopsin Comparati6e Physiology A, 154, 153–156. content. In: The biology of photoreceptors, Proceedings of the Zimmerman, W. F., & Goldsmith, T. H. (1971). Photosensitivity of 36th Symposium of the Society for Experimental Biology, 471– the circadian rhythm and of visual receptors in carotenoid-de- 495. pleted Drosophila. Science, 171(976), 1167–1169.