Molecular stabilization and complexation: the secrets of making a nuclear-selective histological stain from naturally occurring anthocyanins without oxidation

ABSTRACT

The natural colorant, roselle, found in Hibiscus sabdariffa, has been used as a histological dye since at least 1976. As a simple extract roselle acts as a general red counterstain, but when treated with an oxidant and metallic mordant it functions as a useful blue nuclear-selective stain. In the past 40 years it has been assumed that oxidation is necessary when preparing a stain from roselle, as of course it is for the related flavonoid dyes hematoxylin and brazilin. However, the chemistry of roselle argues against this. Roselle is a mixture of four closely related compounds: delphinidin 3-sambubioside and cyanidin 3-sambubioside, delphinidin 3-glucoside and cyanidin 3-glucoside. Each of these in turn can exist in 8 different configurations in a complex state of equilibrium largely dependent upon pH. Of these compounds, only three are colored. In plants, and in their complex extracts, formation of these colorless compounds is inhibited by the presence of other colorless aromatic molecules, a stabilizing process termed co-pigmentation. The color of such extracts may be red in the acid range and blue or violet above neutrality. Complexation (chelation) with metal cations occurs above pH 2.8 or 3.0, depending upon the concentration of the metal. Prior oxidation is not needed for such chelation to occur. With Al³⁺ the resulting metal complex, which we term rosalum, is violet or blue. Following spectrophotometric investigation of this complexation, we were able to formulate a nuclear-selective violet/blue stain without the use of an oxidant.

KEYWORDS:

• Anthocyanins: complexation
• co-pigmentation
• nuclear stain
• roselle

Acknowledgments

We thank Dr. Wathiq Sattar Abdul-Hussan, University of Thi Qar, Nasiriyah, Iraq, for technical assistance; Sheila Criswell, University of Tennessee Health Science Center, Memphis, TN; Carol Bain, Indiana University-Purdue University, Indianapolis, IN for help with obtaining specimens; and Richard W. Horobin for scientific advice.

Disclosure statement

No conflict of interest was reported by the authors.