What is the true origin of the bright red-orange autofluorescence in the hepatocytes?


  • Potential conflict of interest: Nothing to report.

What is the True Origin of the Bright Red-Orange Autofluorescence in the Hepatocytes?

To the Editor:

We read with interest the article titled “Copper-Metallothionein Autofluorescence” by Quaglia et al.1 The authors mentioned: (1) fluorescence microscopy is based on the interactions between light wavelengths, molecules, and filters/dichromatic mirrors, which may unmask specific combinations for the identification of particular structures; (2) autofluorescence-based techniques have been used for the assessment of fibrosis, cell metabolism, and differentiation; discrimination between neoplastic and non-neoplastic tissue; and studies of microorganisms and drug interaction. We agree with the authors' descriptions concerning fluorescence microscopic techniques. However, we should usually recognize that these factors are restrictive and exceptional. Here, we would like to demonstrate that there was another possibility in their observations.

Their fluorescence microscopy systems were as follows: dichromatic mirror reflecting at 415 nm wavelength, excitation filter at 420 ± 30 nm wavelength, and suppression filter at 465 ± 20 nm wavelength. The unstained 6-μm-thick sections were excited at the regions between 390 nm and 415 nm. These regions did not match for autofluorescence from copper-metallothioneins (minutely cuprous [Cu(I) or Cu+] metallothioneins [MTs]), because a maximum peak of excitation wavelength for Cu(I)-MTs is around 300 nm.2 Therefore, it can be imagined that the emission from Cu(I)-MTs was very weak. Mysteriously, the red-orange autofluorescence, which originated from Cu(I)-MTs at around 600 nm,2 was observed when illuminated with the excitation filter, although the suppression filter at 465 ± 20 nm shades the light of wavelength more than 485 nm. Moreover, Cu(I)-MTs were identified with the comparable colocalizations of the red-orange light and of the staining spots with rhodanine and orcein, although these chemicals bind basically with Cu2+ but not with Cu+.3 Therefore, the above conditions were insufficient proof concerning the red-orange autofluorescence from Cu(I)-MTs.

We indicate that the best filter set for the fluorescence microscopic observations of Cu(I)-MTs is a dichromatic mirror at 400 nm, excitation filter at 330-385 nm, and barrier filter at 420 nm because they emit the most strongly when specimens are illuminated with excitation in the 280-350 nm region.2 Using this filter set, bright yellow-orange autofluorescence was observed in the livers of the Long-Evans Cinnamon (LEC) rats (an animal model of Wilson's disease) just before spontaneous acute hepatitis (at the age of 15 weeks).4 The autofluorescence was diffuse in the cytoplasm of randomly distributed hepatic parenchymal cells (Fig. 1). The emission was observed on some vacuolated nuclei of hepatocytes, and in spherical granules of various sizes and densities in some hepatocytes and in Kupffer cells. All the emissions were present in the periportal zone and midzone of liver lobules, but not in the centrilobular zone, and were absent in the epithelial cells of hepatic veins, arteries, and bile ducts.4

Figure 1.

Fluorescence microscopic photograph of bright yelloworange autofluorescence from Cu(I)-MTs in hepatic sections of a 15-week-old male Long-Evans Cinnamon (LEC) rat. The hepatic tissues were cut in a cryostat microtome (Model CM-41; Sakura Seiki Inc., Tokyo, Japan) at –20°C. The frozen sections (8 μm) were mounted on slides, thawed at room temperature, and immersed in acetone for 5 minutes to dehydrate the tissues.6, 7 The specimens were embedded rapidly in Entellan new (Merck, Darmstadt, Germany).6, 7 These observations were performed with the aid of an ordinary fluorescence microscope, an Olympus BX50-FLA microscope (Olympus, Tokyo, Japan) equipped with a 100-W mercury lamp for epifluorescence6, 7 and with a fluorescent digital camera (Olympus DP72; Olympus, Tokyo, Japan). The U-MWU filter set was used (a 400 nm wavelength dichromatic mirror; a 330-385 nm excitation filter; and a 420 nm barrier filter). Magnification, ×50. Bar = 50 μm.

So, what was the true origin of the bright red-orange autofluorescence in the report by Quaglia et al.? There are two possible solutions. The first is that the excitation regions between 390 nm and 415 nm are the best for autofluorescence from porphyrins because porphyrins emit bright red-orange when they are excited in Soret's band around 405 nm.5 Actually, we established by using microspectrophotometry that the red-orange autofluorescence in 30-week-old male LEC rat kidneys was from the emission of porphyrins.6, 7 The second hypothesis is that there are many articles about red-orange autofluorescence in hepatocytes with liver disease, such as hepatitis, liver cirrhosis, porphyria cutanea tarda, and especially hepatocellular carcinoma. However, most reports were published from the 1950s to the 1980s.8-10 Unfortunately, we cannot see the precious color photographs of the red-orange autofluorescence from porphyrins in those livers, because most of those published photographs were black and white. Therefore, it has been forgotten that the origin of the red-orange autofluorescence in the liver tissues was from porphyrins.

We believe that the truth is usually simple and obvious. We assert that there are phenomena in which both porphyrins and Cu(I)-MTs are colocalized in the cells of liver and/or kidneys. Those who detect autofluorescence with a red-orange and/or yellow-orange color in the cells should not focus only on the color, because our eyes cannot analyze and calculate the wavelengths. No one has ever confirmed biomaterials by watching the emitting color.

How long will the debate between autofluorescence arising from porphyrins and that arising from Cu(I)-MTs continue? This unresolved conflict results in lost time, money, and human lives. Therefore, we wish to emphasize the importance of examining the spectral properties of the emissions to avoid any serious mistakes such as confusing porphyrins with Cu(I)-MTs.


This study was supported by grants-in-aid for Scientific Research from the Japan Society for the Promotion of Science (19590658). We are indebted to Professors N. Kasai (Institute for Animal Experimentation, Graduate School of Medicine, Tohoku University), T. Ohyama (Department of Food Science and Technology, Faculty of Bioindustry, Tokyo University of Agriculture), and Drs. K. Jin (Hokkaido Institute of Public Heath) and T. Okui (School of Veterinary Medicine, Rakuno Gakuen University), and Mr. H. Yamada (Olympus, Sapporo, Japan) for their support during this study. We are grateful to our colleagues, especially F. Takenaka and H. Mikami for their helpful animal management.

Kenji Nakayama Ph.D.*, Mamoru Tamura Ph.D.†, * Department of Health and Environmental Science Hokkaido Institute of Public Health, Sapporo Hokkaido, Japan, † School of Medicine, Tsinghua University Haidian District, Beijing, China.