- Top of page
- Materials and methods
Sickle cell disease (SCD) is characterized by vasculopathy, which has been causally linked to intravascular haemolysis and high levels of free plasma haemoglobin. Soluble CD163 (sCD163) is implicated in the clearance of free plasma haemoglobin and high plasma concentrations have been linked to arterial disease. We therefore investigated the value of sCD163 as a biomarker in children with SCD, and also measured haptoglobin levels in this population. We measured sCD163 in 25 control children with no haemoglobinopathy, 41 with sickle cell anaemia (HbSS) in the steady state, 27 with HbSS taking hydroxycarbamide, and 7 with HbSC disease. There was no significant difference between sCD163 levels in steady-state HbSS (1·78 mg/l) and controls (1·81 mg/l) (P = 0·86). However, sCD163 levels were significantly lower in those HbSS children taking hydroxycarbamide (1·35 mg/l) compared to both steady state HbSS (P = 0·004) and controls (P = 0·036). In children on hydroxycarbamide, sCD163 correlated negatively and highly significantly with percentage HbF (R = −0·76, P < 0·001), and this relationship was absent in those not taking hydroxycarbamide (R = 0·07, P = 0·65). sCD163 is a potentially useful biomarker in children with SCD, and may have a role in monitoring responses to hydroxycarbamide.
CD163 is a member of the scavenger-receptor cysteine-rich family of proteins. It is expressed on cells of the monocyte-macrophage lineage and is the main haemoglobin-haptoglobin (HbHp) receptor (Kristiansen et al, 2001). In addition to clearing haemoglobin from the plasma, it mediates the interaction between macrophages and erythroblasts (Fabriek et al, 2007), is a TWEAK [TNF (tumour necrosis factor)-like weak inducer of apoptosis] receptor (Bover et al, 2007), acts as a direct receptor for bacteria (Fabriek et al, 2009) and is an immunomodulator (Van Gorp et al, 2010). CD163 is actively shed from the surface of monocytes/macrophages and soluble CD163 (sCD163) occurs at high levels of up to 2 mg/l in normal human plasma (Moller et al, 2002). The functions of sCD163 are unknown, although it has been suggested that it might bind HbHp complexes and reduce cellular iron uptake (Weaver et al, 2006) and has also been shown to inhibit T-lymphocyte activation and proliferation (Hogger & Sorg, 2001). Plasma levels of sCD163 have been investigated as potential markers of inflammation and macrophage activation, and show marked increases in some conditions, such as Gaucher disease (Moller et al, 2004), sepsis (Moller et al, 2006), liver disease (Moller et al, 2007) and peripheral arterial disease (Moreno et al, 2010).
sCD163 is highly relevant when considering the pathophysiology and complications of SCD. There is a significant chronic inflammatory component in SCD due to continuous ischaemic tissue damage, with more inflammation occurring during acute complications (Belcher et al, 2003); sCD163 may therefore be higher in children with more severe disease and predictive of poor outcome. Haemolysis is an important component of the pathophysiology of SCD, causing the release of free haemoglobin in to the plasma, where it binds strongly to and inactivates nitric oxide (NO) (Reiter et al, 2002); this functional NO deficiency causes endothelial dysfunction, which may contribute to progressive vascular damage. There is evidence that the rate of haemolysis, as measured by plasma lactate dehydrogenase (LDH) levels, correlates with some complications in SCD, particularly pulmonary hypertension (Gladwin et al, 2004), leg ulcers, priapism (Kato et al, 2006) and cerebral vasculopathy (O’Driscoll et al, 2008). However, there is very little evidence concerning differences in the rate of clearance of free haemoglobin from the plasma, which is also likely to be a determinant of any damage caused by haemolysis. sCD163 levels may indicate variability in inflammation and the clearance of free plasma haemoglobin, and explain some of the clinical variability which characterizes SCD.
We have therefore measured the sCD163 level in children with SCD and controls to investigate whether it might be a useful marker of disease activity, the effects of treatment with hydroxycarbamide (HC) and its relationship to any complications.
- Top of page
- Materials and methods
Haemolysis and free plasma haemoglobin are increasingly implicated in the vasculopathy of SCD (Kato et al, 2006), although whether their action is mediated through NO is controversial (Bunn et al, 2010). It is therefore important to understand how the clearance of free plasma haemoglobin occurs in SCD, and whether variability in this is responsible for some of the phenotypic differences. Although the clearance of plasma haemoglobin is partly understood in health, there may be significant differences in SCD due to the increased rate of haemolysis, chronic inflammation and hyposplenism. CD163 is central to this process, being the main HbHp receptor. The measurement of sCD163 in the plasma of children with SCD has not previously been reported and has the potential to shed light on haemolysis-related endothelial dysfunction; for example, high sCD163 levels have been found to be a marker for coronary atherosclerosis (Aristoteli et al, 2006).
We did not find any relationship between the rate of haemolysis, haemolytic complications and sCD163 concentrations. For example, there was no significant difference between levels found in steady-state HbSS and healthy controls. Similarly, there was no correlation between sCD163 levels and markers of haemolysis, such as LDH or vasculopathy, as indicated by transcranial Doppler scan velocities. This lack of association suggests that sCD163 does not play an important functional role in binding to HbHp complexes in chronic haemolysis, as has also been suggested by in vitro studies showing that sCD163 binds HbHp complexes weakly compared to the membrane-bound form (Moller et al, 2010). However, it is unclear from our study whether the monocyte-macrophage expression of CD163 might vary in SCD and contribute to phenotypic differences; this possibility is indirectly supported by evidence that corticosteroids increase the expression of CD163 on monocytes (Schaer et al, 2002) and are also of therapeutic benefit in acute pain (Griffin et al, 1994) and acute chest syndrome (Bernini et al, 1998). It is also possible that although sCD163 levels were normal in SCD, this is a balance between both increased production and increased removal by HbHp complexes.
Surprisingly our study showed that HC significantly reduced concentrations of sCD163 in children with SCD, below that found in the steady-state and also below concentrations in paediatric controls. This decrease showed a strong correlation with increasing HbF levels, but no correlation with MCV. It is possible that the fall in sCD163 is related to the myelosuppressive effects of HC, being directly linked to the reduced numbers of monocytes and macrophages. There was no significant difference in monocyte numbers between those taking and not taking HC (0·79 × 109/l vs. 0·99, P = 0·09), although there was a trend towards lower monocytes in those on HC, and an effect on resident macrophages that are supposed to be the main source of sCD163 cannot be excluded. It is possible that the higher HbF levels result in less chronic inflammation and vascular damage, and that the lower sCD163 levels are indicative of this. An alternative explanation is that HC has a direct effect on sCD163 concentration that is proportionate to its effects on HbF levels; this could be caused either by direct suppression of sCD163 production or reduced rates of release from monocyte/macrophage membranes. Another possibility is that HC directly nitrosylates the cysteine-rich sCD163 molecule, causing its degradation. sCD163 has the potential to be a useful biomarker to monitor HC therapy, and may also shed further light on mechanisms of action of HC in SCD, including its role in promoting HbF. Future studies could address its prognostic significance in comparison with HbF.
Using conventional assays, haptoglobin is undetectable in 75% of patients with SCD (Podmore et al, 2007). However, using this highly sensitive assay, haptoglobin was measurable in all our patients, but did not correlate with any known markers of haemolysis, such as LDH, or other clinical events. As expected, haptoglobin levels were lower than normal in HbSS, and to a lesser extent in HbSC. There was no significant difference in haptoglobin levels between those taking and not taking HC. However, haptoglobin levels were detectable using this assay, suggesting that the HbHp system continues to offer some protection in SCD against the damaging effects of free plasma haemoglobin. It is, however, possible that the low Hp levels determined in patients with SCD partly reflect haptoglobin-related protein due to cross reactivity of the antibody used in the ELISA. We did not examine the role of haptoglobin polymorphisms, haptoglobin-related-protein or haemopexin in this study.
Our study suggests that sCD163 is a potentially useful biomarker in SCD, and may be particularly relevant in monitoring the effects of HC. The significant correlation between HbF percentages and sCD163 concentrations in children on HC may lead to further insights into the mechanisms of action of HC. Further studies of sCD163 and haemoglobin scavenging in SCD are likely to be informative.