337Oxygen Dissociation Assay (ODA): spectrophotometric based screening platform for hemoglobin-O2 affinity modifiers

Mira P Patel (1), Vincent Siu (1), Abel Silva-Garcia (1), Qing Xu (2), Zhe Li (2), Donna Oksenberg (1), (1) Biology Department, (2) Chemistry Department, Global Blood Therapeutics Inc., South San Francisco, CA, 06/2019
  • A novel screening assay named the oxygen dissociation assay (ODA) was developed
  • Assay is based on the spectral changes observed during hemoglobin deoxygenation
  • SPECTROstar Nano captured spectral changes over time during N2 addition to microplate chamber


Hemoglobin (Hb) is an iron-containing globular protein present in the blood cells of all vertebrates. Its ability to perform transport of gases, primarily O2, make it critically important for maintaining aerobic metabolism.

Compounds which exogenously modify Hb affinity for O2 are desirable for treating multiple diseases. Modifiers that decrease Hb-O2 affinity would be useful when improved delivery of O2 is needed, such as in wound healing. In addition, increased Hb-O2 affinity would be useful in sickle cell anemia to delay the polymerization of deoxygenated sickle Hb.

Here we describe the Oxygen Dissociation Assay (ODA). Relatively high throughput was obtained using the SPECTROstar Nano to read 96-well plates. The reader was equipped with an optional gas-vent so that N2 could be directly added to the reader and spectral data collected over time to monitor the deoxygenation of Hb.


Assay Principle

To determine the O2 affinity of Hb the spectral differences between oxy- and deoxy-Hb were employed (Figure 1).

Fig. 1: ODA Assay Principle. The absorbance spectra of oxygenated and deoxygenated hemoglobin differ and are used to determine Hb oxygen affinity.

Differences between the 2 forms of Hb are primarily seen at the Soret band (400-500 nm) and the Q-band (500-600 nm) but the entire spectrum from 350 to 700 nm was monitored every 6 minutes for 2 hours while N2 was flowing through the microplate reader. The spectral data were analyzed using Excel’s LINEST function and converted to % of oxygenated Hb.

Materials & Methods

  • 96-well, half-area microplate (µ-clear bottom, Greiner)
  • SPECTROstar Nano equipped with gas vent
  • For a complete description of reagents and sources see Patel et al.1

Experimental Procedure
3 µM purified Hb was added to plates with or without various compounds. Plates were sealed and incubated at ambient air condition for 1 hour at 37°C. Plates were subsequently unsealed and placed in the SPECTROstar Nano to which gaseous dry N2 was applied via the gas vent. The plate was read in the following way:

Instrument settings


Optic settings

Absorbance, plate mode kinetic 

Wavelength range:

350-770 nm

General settings


Number of flashes:


Settling time:

0.2 s

Kinetic settings


Number of cycles:


Cycle time:

360 s

Shaking settings



Shaking frequency:

300 rpm

Shaking mode:

Double orbital

Shaking time:

60 s - each cycle




Results & Discussion

Figure 2 compares the spectral results for Hb at the beginning and end of the 2 hour exposure to N2. There is a clear rightward shift in the peak of the Soret band from 415 to 430 over the course of the experiment. Furthermore the 2 peaks at the Q band (541 and 577 nm) at the beginning of the exposure, resolve into one peak (555 nm) at the end of the 2 hours.

Fig. 2: Spectra of Hemoglobin over 2 hours of deoxygenation. Overlay of spectra collected for Hb at t = 0 (red) and t = 120 minutes (purple)

The 20 spectra that were collected over time were compared across the wavelength range from 380 to 700 nm and the results were expressed as %oxy Hb. The %oxy Hb results were plotted against time (Figure 3) showing the expected decrease in %oxy Hb over time.

Fig. 3: Relationship of %oxy Hb and time. Each of the 20 spectra collected (n = 3) during the 2-hour experiment were analyzed using Excel’s LINEST function. The results plotted vs. time allow for %oxyHb quantitation. Patel et al. Drug Design, Development and Therapy 2018:12 1599-1601. Originally published by and used with permission from Dove Medical Press Ltd.

The ODA was then tested in the presence of well characterized allosteric modifiers of Hb-O2 affinity. Figure 4 shows a comparison of ODA results with Hb alone and in the presence of phytic acid and GBT1118.

Fig. 4: The %oxy Hb over time +/- affi nity modifi ers. In the presence of GBT1118 the oxygenated state of Hb is stabilized while phytic acid sustains the deoxygenated state. Patel et al. Drug Design, Development and Therapy 2018:12 1599-1601. Originally published by and used with permission from Dove Medical Press Ltd.

Analyzing oxygen equilibrium curves (OEC) with the industry standard TCS Hemox Analyzer in presence of GBT1118 and phytic acid reveals differences in Hb oxygenation (data not shown). A more extensive ODA screen was compared to OEC results and it was found that the compounds that decrease oxygen affinity of hemoglobin, like phytic acid, are best evaluated at 30 minutes in the ODA. These compounds shift the ODA curve to the left due to low affinity and fast displacement of oxygen. Whereas, compounds that increase oxygen affinity of hemoglobin shift the ODA curve to the right as oxygen dissociation is delayed. These compounds such as GBT1118 are best evaluated at 108 min. These 2 detection times are indicated in figure 4. Regardless of the evaluation time used the results of ODA correlated well with OEC and exhibited acceptable robustness statistics (data not shown).


The ODA represents a major step toward HTS in analysis of Hb-affinity modifiers. Compared to industry standards the ODA improves upon both the time required and volume required for the screening assay.


  1. Patel, MP et al. Development and validation of an oxygen dissociation assay, a screening platform for discovering, and characterizing hemoglobin-oxygen affinity modifiers. Drug Des. Devel. Ther. (2018) 12: 1599-1607.
Newsletter Sign-up