Simultaneous Determination of Free Heme and Free Hemoglobin in Biological Samples By Spectral Deconvolution

Introduction: In recent years, free heme, as well as free hemoglobin, have been independently identified as key components in the pathophysiology of many different disease processes, e.g. sickle cell anemia, sepsis, and trauma. Accordingly, it is imperative that methods used to measure both free heme and free hemoglobin are accurate, precise, and replicable. In vivo, free heme and free hemoglobin exist together and must be distinguished from each other when attempting to quantitate. Methods described previously to quantify free heme involve the use of manufacturer colorometric assay kits. However, these assays typically cannot distinguish between free heme and free hemoglobin which necessitates their separation before measurement. Separation may be achieved by filtration on the basis of molecular weight differences. Whilst this approach has been used, we speculate that it underestimates free heme levels due to the latter’s hydrophobicity and entrapment in filtration membranes.

Objective: The first objective of this study is to determine if previous methods which have utilized a Centricon filtration device to obtain measurements of free heme and free hemoglobin are accurate. The second objective will look at the potential use of a desalting column to separate free hemoglobin from free heme. The third objective is to develop a new method, based on spectral deconvolution, that would allow for rapid and simultaneous measurement of both free heme and free hemoglobin in biological matrices

Methods: Standard solutions of free heme, free hemoglobin, or mixtures of both were quantified using commercially available colorometric assay kits specified for selective measurement of each species. In addition, mixtures of free heme and hemoglobin were separated from each other prior to measurement using two different techniques. The first utilized a centrifugal filter which traps molecules greater than 10,000 daltons and the second, a desalting column which retains molecules less than 6kDa.

We developed a spectral deconvolution approach to simultaneously measure the concentrations of oxyhemoglobin, methemoglobin and free heme in solution. This approach relies on individual species within a complex mixture having distinct absorbance spectra, and availability of spectra for a known concentration of each species alone. Concentrations of each individual species within a mixture are determined by deconvolution of experimental spectra against standards by multi-linear regression fitting. This method was validated by using cyanide which selectively ligates and changes the spectrum of metHb only.

Results: Both kits tested were able to detect free heme or hemoglobin with equal sensitivity, and neither was able to distinguish between free heme from free hemoglobin. Utilization of the Centricon filtration system should have trapped free hemoglobin while allowing the free heme to filter through; however, this data shows that all of the free heme was also retained leading to a gross underestimation of the amount of free heme as well as an over estimation of free hemoglobin. We posit that this result may be secondary to free heme’s hydrophobicity leading to membrane entrapment. The desalting column was used in an effort to retain the free heme while allowing the free hemoglobin to filter through. This method effectively separated free heme and hemoglobin and allowed for a >85% recovery of free hemoglobin. A standard spectral deconvolution algorithm was created and was able to distinguish between free heme, oxyHb, metHb in complex mixtures with >95% specificity.

Conclusions: As more data emerges showing that both free heme and free hemoglobin are important factors in many different disease pathways, the accurate measurement of both molecules has become vital to understanding these diseases. The data here shows that previously described methods of quantifying free heme and free hemoglobin may not be specific and therefore cannot distinguish between the two molecules. Moreover, approaches to separate heme from free hemoglobin need to be validated and ensure loss of free heme during sample processing is not occurring. We forward a novel method that requires minimal sample handling, and is able to simultaneously quantify free heme and hemoglobin in complex mixtures by spectral deconvolution. We posit that this method may allow for rapid, and accurate measurement of free heme and free hemoglobin in biological fluids.