Vitamin A deprivation does not decrease fluorescence of ARF72-RFP, a label for Golgi apparatus, in Drosophila visual receptors.

Denny, George and William S. Stark. Department of Biology, Saint Louis University. St. Louis, MO 63103. e-mail:

Each ommatidium of the Drosophila compound eye has 8 photoreceptors (retinula cells); the rhabdomere of each is the specialized organelle that houses rhodopsin and the visual transduction molecules. One of the primary functions of the retinula cell is maintenance of the rhabdomere, including turnover of membrane and protein (Lee et al. 1996). Vitamin A deprivation reduces or eliminates rhodopsin in Drosophila rhabdomeres (Harris et al., 1977). Vitamin A replacement synchronizes de novo synthesis and export of rhodopsin (Sapp et al., 1991).

Although the quality of fixation of Drosophila photoreceptors has always been variable, electron micrographs from our lab archives showed that vitamin A deprived flies are more likely to be plagued by what we refer to as ³ghosty cytoplasm² (marked with asterisks [*], Fig., Top left) than vitamin A replete controls (Fig., Middle left). (Electron micrographs are labeled thus: R is rhabdomere, G is Golgi, > is desmosome, N is nucleus, and PG is pigment granule.) We hypothesized that cytoplasmic organelles dedicated to biosynthesis, rough endoplasmic reticulum and Golgi apparatus, might be reduced by vitamin A deprivation. We tested our hypothesis using a fly stock we had been using to visualize Golgi apparatus, ARF72-RFP (ADP ribosylation factor tagged with red fluorescent protein).

Flies were lightly etherized and fixed to a glass slide for visualization of the deep pseudopupil. A typical fluorescence micrograph is presented (Fig., Top right); the blurry appearance compared with rhabdomere fluorescence (Stark and Thomas, 2004) is explained since Golgi apparatus is distributed throughout retinula cells. Fluorescence was quantified using a fluorescence microscope with a photometer system (Stark et al. 1985). The pseudopupil image was delimited by the photometer and fed to the photomultiplier tube. Rhabdomeres were excited with a calibrated amount of 488 nm light, and a voltage response proportional to the level of fluorescence being emitted was recorded by a computer. Flies were raised at room temperature either on our yellow corn meal food (supplemented with beta-carotene, vitamin A replete) or on Sang¹s medium lacking vitamin A. A strong correlation between age and level of fluorescence was noted (Fig., Middle right). Thus, for control, measurements were performed on flies within 12 hr of eclosion.

RFP fluorescence of vitamin A deprived flies (n=28) was the same as that of flies reared on vitamin A replete medium (n=27), as witnessed since the error bars (95% confidence intervals) overlap (Fig., bottom left). The stock of flies was not genetically pure; only 28.7% of the ARF72-RFP flies raised on the replete medium and 18.3% of the flies raised on the deficient food showed RFP fluorescence. Furthermore, we noted a variation in eye color – ranging from slightly pink to pure white (in the white-eyed stock). However, there was no correlation found between eye color and whether a fly showed RFP fluorescence. Measuring at a higher sensitivity to quantify background autofluorescence among flies that were negative for RFP, the level was higher in vitamin A deprived flies (n=125) than in those raised on replete food (n=67) (Fig. Bottom right); however, again, this difference was not statistically significant as witnessed by the overlapping 95% confidence intervals. The purpose of this last control was to verify the expectation that background fluorescence did not predominate in our measurements of RFP fluortescence.

In conclusion, we reject our hypothesis that Golgi apparatus, as quantified by ARF72-RFP fluorescence, is higher in vitamin A replete Drosophila than in vitamin A deprived flies; further, we conclude that any difference in quality of fixation between replete vs deprived Drosophila, if real, cannot be attributed to different amounts of Golgi complexes in retinula cells.

Acknowledgements: We thank Prof. J.E. O¹Tousa at University of Notre Dame for the white-eyed ARF72-RFP stock.

            References: Harris, W. A., et al., 1977 Nature (Lond.) 266: 648-650; Lee, R. D. et al. 1996, Micros. Res. Tech. 35, 418-430; Sapp, R. J. et al., 1991 Exp. Eye Res. 53: 73-79; Stark, W. S. et al., 1985 Curr. Eye Res. 4: 1059-1075; Stark, W. S., and C. F. Thomas 2004, Molec. Vision 10: 943-955 on line at