Background: Under hypoxic conditions, sickle hemoglobin (HbS) polymerizes, causing morphologic distortion (sickling) of red blood cells (RBCs) in sickle cell disease (SCD). Aes-103 (5-hydroxymethylfurfural, 5-HMF) can stabilize the R-state and increase the oxygen affinity of hemoglobin, inhibiting the intracellular polymerization of HbS. Using a microfluidics-based hypoxia assay, we were able to track sickling of individual cells and quantify the anti-sickling effect of Aes-103 at millimolar (mM) levels in blood from SCD patients on hydroxyurea treatment (on-HU) and not on hydroxyurea treatment (off-HU).

Method: We have developed a microfluidic assay that utilizes a gas permeable polydimethylsiloxane (PDMS) film 150 µm in thickness, to create a severe hypoxia microenvironment in a 5 µm deep chamber to measure cell sickling in vitro at 37°C. The hypoxia condition was 5 minutes in total, consisting of an initial oxygen-rich stage (20% O2), a transient deoxygenating stage (O2 concentration decreased to 5% within 15 second), and a steady-stage stage (O2 concentration decreased further and maintained at 2% for the rest of time). Blood samples from 3 on-HU and 3 off-HU patients were incubated with Aes-103 at concentrations of 0.5, 1, 2, and 5 mM for one hour at 37 degrees C, washed with Phosphate Buffered Saline and suspended in RPMI-1640 containing 1% w/v Bovine Serum Albumin for in vitro testing. Sickle RBCs undergoing sickling typically form spiky edges and a dark coarse texture due to intracellular HbS polymerization visually enhanced by a bandpass filter (Fig. 1A). The anti-sickling effect of Aes-103 was then quantified by the maximum sickled fraction (fraction of all RBCs that were morphologically distorted) under the hypoxia condition.

Results: In the absence of Aes-103, the sickled fractions varied from 34% to 73% (Mean ± SD: 54% ± 18%). With the presence of Aes-103, the mean sickled fraction decreased with drug concentration (Fig. 1B), which can be well fitted with linear regression (R2= 0.95). With 2 mM Aes-103 incubation, each patient sample showed a significant decrease in cell sickling from its baseline. Addition of Aes-103 at 5 mM concentration prevented majority of RBCs from sickling (sickled fraction ≤ 5%). The sickled fraction of one patient sample was nearly zero. The distribution of sickled fractions does not completely correlate with the patient's HU status in this limited sample size (Fig. 1C). We also observed that hypoxia-induced sickling at baseline showed an apparent bimodal distribution, although the slope of response to Aes-103 concentration was similar.

Conclusions: Our microfluidic assay enabled a rapid, quantitative characterization of cell sickling in vitro within a few minutes and using a single drop of whole blood patient sample. We confirmed the anti-sickling efficacy of Aes-103 for both on-HU and off-HU patient samples in a dosage-dependent manner. This assay has potential as a biomarker for drug development and monitoring for in vivo effect of potential anti-sickling therapeutics.

Figure 1. (A) Identification of cell sickling from a microscopic image (arrows indicate the sickled RBCs). (B) Sickled fraction as a function of Aes-103 concentration. (C) Variation in response among different on-HU and off-HU patient samples.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.