Quantitative cytochemistry of glycogen in blood cells. Methods and clinical application

Abstract

Quantitative glycogen determinations can be made in single blood and bone marrow cells, using microspectrophotometry or microfluorometry after staining with variants of the periodic acid--Schiff (PAS) reaction. These PAS variant reactions generally do not indicate the presence of non-glycogen PAS-positive substances, known to be prevalent in various hematopoietic cells, possibly due to masking of reactive groups. The specificity of the reaction in blood cells was ascertained by alpha-amylase digestion, which removed more than 95% of the PAS-positive material. Calibration of the PAS reaction was undertaken with a microdroplet model of pure leukocyte glycogen. The glycogen amounts in the droplets were determined by microinterferometry, the droplets were stained with a variant PAS reaction, and the total extinction of the reaction product in the stained droplets was determined by microspectrophotometry. The extinction coefficient (k) was obtained from the equation k equals Etot divided by M where (Etot) is the total extinction as determined by microspectrophotometry and (M) the dry glycogen amount as determined by microinterferometry. The microinterferometric dry mass determinations were calibrated by X-ray absorption in order to obtain the absolute amounts of glycogen. For practical purposes a reference system was made of normal neutrophil leukocytes. The glycogen content in the reference neutrophils was first determined with the micromodel. These neutrophils, now with a known glycogen amount, were stained with the PAS reagents and measured microspectrophotometrically in parallel with cells containing an unknown glycogen amount. Alternatively, the staining was made with a fluorescent PAS reaction, and the glycogen content determined by microfluorometry. Both methods appeared suitable for determining the glycogen content of blood cells from patients with various diseases, though the microfluorometric method was preferable for measurements of small amounts of inhomogeneously distributed glycogen. The mean glycogen content of normal neutrophil leukocytes was found to be 13.6 times 10(-12) g. The content was increased in infectious diseases such as pneumonia and tonisillitis, as well as in polycythemia vera and myelofibrosis, while low amounts were found in untreated chronic myelocytic leukemia. In chronic myelocytic leukemia in remission, the glycogen content of mature neutrophils had completely normalized. Erythroblasts normally do not contain detectable amounts of glycogen. However, in certain diseases such as beta-thalassemia and Di Guglielomo's syndrome, appreciable amounts of glycogen accumulate in the erythropoietic precursor cells. In beta-thalassemia this was associated with an arrest in the proliferation of early polychromatic erythroblasts, which accumulate glycogen in the G1 phase of the cell cycle. In all these diseases quantitative glycogen determinations in the blood cells have diagnostic importance.