Rapid and accurate heme quantitation in the research lab has become more desirable as the crucial role that intracellular hemoproteins play in metabolism continues to emerge. Here, the time-honored approaches of pyridine hemochromogen and fluorescence heme assays are compared with direct absorbance-based technologies using the CLARiTY spectrophotometer. All samples tested with these methods were rich in hemoglobin-associated heme, including buffered hemoglobin standards, whole blood from mice, and murine erythroleukemia (MEL) and K562 cells. While the pyridine hemochromogen assay demonstrated the greatest linear range of heme detection, all 3 methods demonstrated similar analytical sensitivities and normalized limits of quantitation of ∼1 µM. Surprisingly, the fluorescence assay was only shown to be distinct in its ability to quantitate extremely small samples. Using the CLARiTY system in combination with pyridine hemochromogen and cell count data, a common hemoglobin extinction coefficient for blood and differentiating MEL and K562 cells of 0.46 µM-1 cm-1 was derived. This value was applied to supplemental experiments designed to measure MEL cell hemoglobinization in response to the addition or removal of factors previously shown to affect heme biosynthesis (e.g., L-glutamine, iron).
Heme is an essential and central component in diverse biological processes ranging from gas metabolism, one-electron chemistry, and bacterial pathogenesis (1-3) to its more recently recognized roles in circadian rhythm and small RNA processing (4,5). From a clinical standpoint, the level of hemoglobin-associated heme in blood and body fluid samples can be diagnostic of many pathological conditions. Here, we have focused on the quantitation of heme in standard hemoglobin solutions, erythroid model cell lines, and blood samples as a benchmark for comparison of currently available technologies.
Traditional strategies for measuring hemoglobin include, but are not limited to, Drabkin's (cyanmethemoglobin) method (6), the pyridine hemochromogen assay (7), the fluorescence heme assay (8), and reverse-phase HPLC (9). With the exception of the cyanmethemoglobin assay, these approaches infer hemoglobin concentrations from measurements of heme. More recently, chemiluminescent-based methods based on horseradish peroxidase biochemistry have also been developed to quantitate total cellular heme (10) and regulatory heme (11). We have previously relied on the pyridine hemochromogen and fluorescence assays to determine cellular heme and hemoglobin concentrations as described elsewhere (12,13). Fluorescence analysis is a more sensitive technique per se, yet signal emission is complicated by multiple environmental factors and therefore does not permit application of the Beer-Lambert law (14). Like the chemiluminescence method, both pyridine hemochromogen and fluorescence assays involve laborious sample preparations that require the handling and disposal of hazardous chemicals.
Two traditional methods of heme quantitation, pyridine hemochromogen and fluorescence heme assays, are compared side-by-side to absorbance spectroscopy using the CLARiTY spectrophotometer. The CLARiTY system offers the advantage of direct measurement of heme in intact cells without the need for derivatization and the concomitant handling of hazardous chemicals. Consequently, this approach allows for post-scan processing (e.g., qPCR, immunoblotting) or reculturing of samples.
As an alternative, the CLARiTY 1000 spectrophotometer by Olis, Inc. (Bogart, GA) has made it possible to measure hemoglobinization in turbid whole-cell suspensions quickly and without sample derivatization. This technology has been implemented in situ elsewhere (15-18), and here we demonstrate use of the CLARiTY system (Figure 1A) to quantitate heme in buffered blood and cell culture resuspensions. The key component of the CLARiTY is a novel integrating cavity absorption meter (ICAM) that contains a quartz cuvette surrounded by a highly reflective coating. UV/visible light bracketing the heme Soret band is generated by a rapid-scanning monochromator (RSM) and directed into the cuvette to produce isotropic, fully diffused light that is subsequently altered only by sample absorption. The light is trapped within the reflective confines of the cuvette until it encounters the output port; thus, its effective pathlength is increased significantly. The enhanced pathlength is inversely related to the detected (apparent) sample absorbance, and this nonlinear effect is corrected by converting the apparent absorbance to absorbance per centimeter using first principles and the Fry Equation (19). The resulting sensitivity of the CLARiTY to light absorption in turbid samples is overwhelmingly superior to that of traditional transmission spectrophotometers (Figure 1B).
Here, we assess and validate our protocol for hemoglobin quantitation using the CLARiTY system in a side-by-side by-side comparison to pyridine hemochromogen and fluorescence heme assays of standard hemoglobin solutions and blood samples. Pyridine hemochromogen and CLARiTY assays are further employed to analyze hemoglobinization in cultures of differentiating MEL and K562 cells. Specifically, we demonstrate the use of heme concentrations from pyridine hemochromogen spectra to calculate red blood cell indices for mouse blood and cell culture samples, a heme proportionality constant for the fluorescence assay, and a micromolar hemoglobin extinction coefficient for the CLARiTY. Additional tests of the CLARiTY verify the ability of our system to determine the effects of oxygen, glutamine, iron, aminolevulinic acid (ALA), and succinylacetone (SA) on hemoglobin levels in MEL cells. In general, we expect the protocol described here to be useful for heme quantitation in many non-erythroid cell types as well. One exception is plant cells, in which the absorbance signatures of chlorophylls and carotenoids significantly overlap the heme Soret peak. Materials and methods Hemoglobin standards and mouse blood
Lyophilized hemoglobin from bovine blood (Sigma-Aldrich, St. Louis, MO) was dissolved in phosphate buffered saline (1× PBS) to a concentration of 20 µM and diluted for use as standards. Whole mouse blood samples were generously provided by L. Wang (University of Georgia, Athens, GA). Samples were collected in K2EDTA tubes from adult wild-type C57BL/6 mice (one female and one male) and subsequently diluted between 1:250 and 1:100,000 in 1× PBS and kept on ice. Cell culture
Cells from the MEL strain DS19 (20) and K562 cell line ATCC CCL-243, originally established by Lozzio and Lozzio (21), were maintained at 37°C and 5% CO2 in complete media consisting of DMEM with 25 mM glucose and 1 mM sodium pyruvate (Cellgro, Corning, NY) supplemented with 2 mM L-glutamine, 9% (v/v) FBS (Atlanta Biologicals, Atlanta, GA), and 1× Pen/Strep (Cellgro). K562 cells were seeded at 1 × 105 cells/mL and induced to differentiate in complete media containing 1 mM sodium butyrate (Sigma-Aldrich). MEL cultures were seeded at 2.5 × 105 cells/mL and induced in media with 1.5% (v/v) DMSO (Sigma-Aldrich). Multi-day induction courses of MEL and K562 cells were carried out in triplicate. In addition, MEL cells were induced for 72 or 96 h under each of the following sets of conditions: (i) 0–4 mM L-glutamine (Cellgro); (ii) 5% and 20% O2; (iii) 100 µM iron as iron dextran (Fisons, UK) and/or 100 µM ALA (Sigma-Aldrich); and (iv) 500 µM SA (Sigma-Aldrich). Hypoxic cultures were grown in a sealed incubator flushed with 5% O2/5% CO2/balance N2 at 37°C. Cell counting
Mouse red blood cells (1:500 dilution in 1× PBS) were counted on a hemocytometer. MEL and K562 cells were washed and diluted in 1× PBS before cell counts were done with a Scepter handheld cell counter (Millipore, Billerica, MA) as previously described (22). MEL and K562 cell proliferation values were obtained by dividing the ratio of live cells at the end of each time course to the number of cells seeded at the beginning by the same ratio for untreated cultures. Spectrophotometric hemoglobin quantitation using the CLARiTY
Absorbances for hemoglobin standards and whole-cell suspensions were measured in the Olis CLARiTY 1000A spectrophotometer with an integrated RSM-ICAM containing a 1-mL cuvette. A total of 10 scans from 280–520 nm were collected per sample. Apparent absorbance values for 1-mL samples were recorded relative to a 1× PBS baseline at the heme Soret peaks of 400 nm for the standard solutions (which contained methemoglobin) and 410 nm for cell samples. Fry correction (19) was carried out to normalize enhanced pathlength values to 1 cm using SpectralWorks software (Olis, Inc.). The Soret peak baseline was then determined for the corrected spectra by linear interpolation. Heme concentrations were obtained from pyridine hemochromogen assays and combined with Fry-corrected absorbance values from the CLARiTY to define a hemoglobin extinction coefficient for mouse blood and MEL and K562 cell cultures. The assumptions that all cellular heme in these samples is bound to hemoglobin and that each hemoglobin tetramer contains four heme molecules were made. The mean corpuscular hemoglobin (MCH) was then calculated in picograms per cell according to the following form of the Beer-Lambert law,