Serum levels of leptin in Nigerian patients with sickle cell anaemia

Homozygous sickle cell anaemia (SCA) also called HbSS, is a form of sickle cell disease that results from single amino acid substitution (Val for Glu) at the sixth position in each of the two beta globin chain of haemoglobin (α₂β^s ₂) and has remained a non-infectious disease of high morbidity and mortality world wide particularly in black populations, Asia and Mediterranean countries [1].

Nigeria has the highest burden of HbSS in the world with ~2% of the population affected by the disease [2]. HbSS produces a myriad of metabolic, haematological, nutritional and clinical effects. Studies have shown that HbSS is associated with decreased food intake, weight gain and delayed puberty especially in females [3].

In the steady state of HbSS, patients are clinically stable but endothelial activation and subclinical vaso-occlusions due to inherent microvascular abnormalities are an ongoing process coupled with chronic but slow red cell haemolysis, platelet activation and moderate cytokine and acute phase response [4]. These pathophysiological events are generally insufficient to induce painful crisis [4, 5].

However, in the unsteady state of sickle cell anaemia (SCA), these pathognomic events are exaggerated and the consequence is severe vaso-occlusion and blood flow that leads to organ damage and painful crisis with complications such as bone and joint infarction, acute chest syndrome, severe anaemia, aplastic anaemia and acute renal failure [6]. These complications have well been reported as the causes of hospitalization and deaths in afflicted patients in Nigeria and other countries [7, 8].

It has been reported that more than 50% of patients with most severe form of HbSS die before the age of 5 years [8, 9]. Studies have also shown that HbSS patients are highly susceptible to infections due to abnormality of their innate immune response including a defective alternate complement system and Plasmodium falciparum parasitaemia has been reported by many investigators as a pathognomic factor of painful crisis in patients residing in malaria endemic countries [10, 11].

In the past 10 years and with the advent of new technologies and discoveries, there has been an increased interest on ways to increase understanding of pathophysiology of HbSS with regards to the roles played by the regulated inflammatory mediators (e.g. α2 macroglobulin), endothelial stress factors (e.g. soluble thrombomodulin) and platelet factors (e.g. beta TG) in the steady state and how these factors can be modified to improve the quality of life and avert painful crisis in patients [12–14].

Leptin, a 16 kDa peptide hormone, derived mainly from the adipose tissue is a prominent biological factor of energy homeostasis whose circulating levels are a reflection of adiposity in human and animals [15]. The hormone binds to the hypothalamus through the leptin receptor and acts via the JAK/STAT pathway to inhibit the expression norexigenic factors but activate the expression of anorexigenic factors to suppress appetite, food intake and weight gain [16]. Therapeutically, leptin has been found to be beneficial in the management of obesity and lipoatrophy in HIV patients [17], while in wasting and decreased energy intake disease conditions such as cancer, tuberculosis, and hepatitis A, B and C infections, decreased serum levels of leptin were found in consonance with decreased cellular immunity [18–21]. In mice, starvation has been used to reduced serum leptin level and evoke an increased susceptibility to endotoxic shock, while in humans uptake of leptin has been associated with an improved T. cell response and intake of micronutrients [22–24]. Leptin has also been found to play important roles in haematopoiesis as it contributes to erythropoiesis and erythropoietin production by the kidney [25]. There are also evidence that leptin potentiates platelet activation [26].

Given the fact that HbSS patients are inadequate in cellular immunity [7–10], hypercatabolic in metabolism [27], prone to anorexia and growth deficit, elicits higher resting energy expenditure (28) and experience chronic haemolytic and inflammatory episodes even in the steady state relative to non-SCD humans [29], there is a need to understand the possible role of leptin in the pathophysiology of HbSS.

The aim of this study was to determine plasma levels of leptin and investigate its associations with body composition, malaria, haemolytic, sickling and inflammatory parameters.

Sickle cell anaemia (SCA) remains a major non-infectious health problem in black populations with Nigeria bearing the greatest brunt of the disease world wide [1, 2, 8, 9]. The results of this study show that haemolytic, sickling and inflammatory episodes coupled with low plasma leptin level before and after correction for fat mass are worsened in Nigeria patients in the unsteady state of SCA.

The low plasma leptin level observed in our SCA cohort agrees with findings of Buchowski [37] who reported a fasting leptin level of 2.26 - 2.36 ng/mL and 5.33 - 7.77 ng/mL in male and female SCD patients relative to 2.49 - 2.59 ng/mL and 15.5 - 28.7 ng/mL in non-SCD male and female controls respectively. After correction for body fat, these workers found that leptin levels in the range of 0.21 -0.74 ng/mL/kg were elicited by both sexes of the SCD patients relative to 0.19 - 1.13 ng/mL/kg by the non-SCD controls. Although similar patterns of reduction in leptin level in SCD patients according to sex were obtained, our observed leptin levels of 2 - 4.1 ng/mL and 2.5 - 12.7 ng/mL for male and female SCD patients are comparatively lower. This disparity might be related to difference in mean ages and sample size of patients used in the two studies. For instance, the mean ages of male and females SCD enrolled in the USA study were 15.4 - 30.6 years and 15.4 - 25.6 years for male and female patients respectively. These ages together with their respective mean weights were higher than those of this study. Although not determined, but considering the geographical divide between the USA study and the present study, dietary pattern of the two enrolled patients may differ and this together with modifiable lifestyle factors have been found to affect the level of leptin in circulation in humans independent of clinical conditions [16, 38]. Similar reduction in leptin level has also been found in patients suffering from fulminant hepatic failure, hepatitis A, B and C infections, non-alcoholic steatohepatitis and tuberculosis [19–21, 39, 40]. To investigate the cause of low plasma leptin level that was further reduced significantly during crisis in our patients, we investigated its relationship with body composition defined by BMI, fat mass. We found no significant correlation between plasma leptin level and both fat mass and BMI, suggesting a possible loss of association between leptin and both adiposity determined by fat mass and BMI in our HbSS cohort. This possibility is further strengthened by our expected finding of positive correlation between leptin and both BMI and fat mass in the non-SCD controls, the observation that the pattern of reduction in plasma leptin levels seen in the two case sub-groups correlated with pattern of reduction in BMI and fat mass by gender and the persistence of lower leptin level in the unsteady SCA sub group after correction for fat mass.

Elsewhere significant positive correlation between leptin and BMI has been found in patients with steatohepatitis, chronic hepatitis B and C infection and acute hepatitis A infections [20, 21, 40], while van Crevel et al [19] and Buckowski et al [37] reported a positive correlation between leptin and fat mass in tuberculosis and sickle cell disease patients. Leptin has been shown in several studies to elicit positive correlation with acute phase proteins and inflammatory parameters such as c-reactive protein, alpha1-antitrypsin, TNF-α, IL-1 and IL-6 [20] to further show the relevance of this hormone as an indicator of inflammation. In this study mean ESR levels of 25.7 mm/hr and 38.4 mm/hr that were above the normal level (i.e. < 15 m/hr) were found in our steady and unsteady patients cohorts and this confirm chronic inflammatory episode aggravated by crisis as a pathophysiologic component of SCD-HbSS. The higher mean ESR level observed in the unsteady SCA patients might be related to the higher parasite rate and parasitaemia seen in these patients since Plasmodium falciparum evoked acute inflammatory response during clinical infection [41]. P falciparum infection detected in the control and steady SCA patients can be said to be sub-clinical and hence asymptomatic. Our finding seems to corroborate a study by Kotila et al [42] who reported an incidence of asymptomatic malaria in SCA patients from Ibadan south West Nigeria. Malaria is transmitted perennially but with high intensity during the rainy season in Nigeria and 98% of the population is at risk of infections [43]. Furthermore, we found lower asymptomatic parasitaemia level of 749.6 parasites/uL among the steady SCA compared to the parasitaemia level of 1078.4 parasites/uL seen in the non-SCD control and this suggest that high level of parasitaemia may be a trigger of unsteady state or crisis in SCA patients. It also indicates a possible lower fever threshold in the study area being deteriorated by crisis in SCA. The relevance of P. falciparum infection to crisis has previously been reported in Nigeria and other SCD population settings where malaria is also endemic [1, 2, 7], while the observed higher pyrexia rate of 58.3% in our unsteady SCA patients compared with 25.8% in the steady patients further supports the possibility of lower fever threshold that we have suggested. Furthermore, several studies conducted in Nigeria and other endemic countries have attributed susceptibility of SCA patients to clinical and severe malaria to poor complement and antibody responses coupled with the ability to enhance parasite sequestration [4, 10].

Since systemic leptin is mainly from the adipose tissue and leptin level rises with increased inflammatory response but decreases with reduced adiposity [15–17], our observed lower plasma leptin level that was exacerbated in an unsteady state in the SCA patients studied represents a balance of these antagonistic events. However, considering the role of leptin in haematopoiesis and effects of adipocyte metabolism on leptin production, the importance of haematological parameters and other adipogenic factors cannot be over-emphasized. The latter was not investigated in the present study, but studies have shown that the levels, uptake or sensitivity to insulin, may influence adipocyte leptin production [44]. With regards to haematological parameters, significantly (P < 0.05) lower level of HbF, a sickling modifier was observed in our unsteady SCA patients compared to their steady counterparts. This suggests that patients in the unsteady state of SCA are poorer in their ability to mount anti-sickling response via HbF and are thus prone to repeated and clinical vaso-occlusions instead of the sub-clinical type, which occurs in a steady state [45].

Under the same shear stress, sickle erythrocytes, owing to their defective cytoskeleton protein-HbS interactions are about 1.5 times and 10 times stickier than normal erythrocyte when oxygenated and during deoxygenation respectively [46]. HbF is known to modify this interaction and its abundance in the erythrocyte has been associated with reduced severity of sickling, vaso-occlusion and clinical manifestations of SCA. This pattern of reduction in HbF level between the steady and unsteady SCA patients studied agrees with previous studies by Ahmed et al [47, 48] who reported HbF levels of 6% and 7% in SCD Nigerian patients without and with priapism in one study and found 5.4% and 5% of this parameter in SCA patients without and with opiate dependence in another study. Therefore, our observed HbF levels of 3.3% and 4.2% in this study is suggestive of greater risk of sickling and vaso-oclussions among our cohorts in Lagos. Unlike in Asia, levels of HbF below 10% are common among African SCA patients including Nigeria [45]. Other abnormal haematological parameters observed in this study such as elevated level of irreversibly sickled cells, mean leukocyte, neutrophils, lymphocyte, monocyte, platelets and reticulocyte counts as well as decreased erythrocyte count, PCV, MCH and MCHC are common haematological features of SCA as they have also documented from several studies for SCA patients in Nigeria and other endemic populations [2, 47–49]. Mild to moderate leukocytosis is common among SCA patients and this has been attributed to re-distribution of leukocytes from marginal pool to a systemic pool of granulocytes, while moderate lymphocytosis and monocytosis have been implicated as markers of inflammation in SCA patients [1, 2, 4, 47, 48]. A study by Wilson et al [50] revealed a positive correlation between leptin level and WBC count, while Togo et al [51] showed that leptin associates negatively with haemoglobin. In this study, we did not find significant correlation between leptin and these haemotological parameters. This is similar to observations by Caner et al [20] in hepatitis A patients and Koc et al [52] in healthy term Japanese infants. Also in this study, the prevalence rates of 29.2% and 79.2% for MCHC < 31 g/dL and MCH < 25 pg were found in our unsteady SCA patients and higher than corresponding rates in steady patients, suggesting higher rate of steady state haemolysis during crisis. Our results are in consonance with similar studies carried out in the northern part of the country and Saudi-Arabia [2, 47–49]. It is also important to know that in this study, the reticulocyte and platelet counts of our unsteady SCA patients were found to be lower than levels in steady SCA group. This is contrary to the findings of Ahmed et al [48] between opiate dependent and non-dependent SCA patients and between tuberculosis and steady SCA patients. However, our results align well with report by the same group of investigators on SCD patients with and without priapism [47] and the findings of Salawu et al [10]. These workers attributed crisis-induced decrease in reticulocyte count to attenuated reticulocyte response to haemolysis in afflicted patients, while crisis-enhanced platelet sequestration was considered to be responsible for their observed platelet reduction. The lowered reticulocyte count in our SCA patients may also be due loss arising from sequestration of the reticulocytes. This is because recticulocytes of SCA patients are more adhesive than those of non-SCD control [2, 4, 5, 7, 8] and the results of this study imply that under unsteady state of SCA, the expression and levels of adhesive molecules factors on the surface of reticulocytes could be more. This is however, subject to further investigations.

Elevated platelet has been associated with inflammation characterized by leukocytosis and elevated ESR in various disease conditions including SCD [2, 4, 47–49] as observed in this study. However, unlike ESR and reticulocyte count, platelet failed as independent predictor of plasma leptin in these patients.

However, this study is limited by the fact that ESR, which was found to be an independent predictor of plasma leptin level is a non-specific marker of inflammation [5] and this may limit its prognostic value in predicting outcome of interventions in SCA patients. Despite this, ESR has been reported as an important inflammatory marker together with other specific markers during steady and unsteady states of SCD [1, 2, 10]. Nevertheless, there may be a need to investigate other more specific inflammatory markers such as β2-macroglobulin and C-reactive protein in our SCA cohort in subsequent studies to build more confidence of reliance on the less expensive and easy to implement ESR assay. Larger sample size is also needed to ascertaining the fitness of poor reticulocyte response as a determinant of low leptin level in SCA. The fact that the lean mass data in this study was obtained by calculation is itself a limitation as it may bias fat mass data of some patients. However, similar calculation was used for all the patients as previously carried out by Owa and Adejuyigbe [31] in the study area to improve the credibility of the corresponding anthropometric data obtained in this study as much as possible. Future studies on the micronutrient variables such as zinc and vitamin E that have been associated with growth and immunity in SCD patients in relation to serum leptin are also advocated.

Nevertheless, within the scope of this study, it can be concluded that unsteady state of SCA is associated with much lower plasma leptin level in relation to inflammation and poor reticulocyte response among Nigerian patients.