Investigation into changes in inflammatory and immune cell markers in pre-diabetic patients from Durban, South Africa

Abstract

The prevalence of pre-diabetes is increasing in rapidly urbanizing cities, especially in individuals aged 25 − 45 years old. Studies also indicate that this condition is associated with aberrant immune responses that are also influenced by environmental factors. This study sought to investigate changes in the concentration of immune cells and select inflammatory markers in patients with pre-diabetes in Durban, South Africa. Blood samples collected from King Edward Hospital, after obtaining ethics approval, were divided into non-diabetic (ND), pre-diabetic (PD) and type 2 diabetic (T2D) using ADA criteria. In each sample, the concentration of immune cells and select inflammatory markers were determined. The results showed a significant increase in eosinophil and basophil levels in the PD group as compared to the ND group. Compared to ND, the PD and T2D groups had significant increases in serum TNFα, CD40L and fibrinogen concentrations. Additionally, there were decreases in serum CRP, IL-6, and P-selectin in the PD group while these markers increased in the T2D group. These findings were indicative of immune activation and highlight the impact of pre-diabetes in this population. More studies are recommended with a higher number of samples that are stratified by gender and represent the gender ratio in the city.

Keywords:

• Immune cells
• inflammatory markers
• pre-diabetes
• Durban
• South Africa

Introduction

Chronic consumption of high-calorie diets has been implicated in the development of insulin resistance in humans and rodents (Arner 2002; Gao et al. 2002; Fonseca 2007; Schrauwen 2007; Myles 2014; Ahmad et al. 2017; Luvuno et al. 2018). Consumption of high-fat diets have been reported to result in increased levels of diacylglycerides that ultimately lead to insulin resistance through activation of protein kinase C (PKC) signaling in the liver and in skeletal muscle (Thompson and Cooney 2000; Bruce et al. 2009; Jornayvaz and Shulman 2012). Prolonged consumption of diets high in carbohydrates have been reported to result in hyperglycemia; concurrently, in these hosts, there is increased activation of nuclear factor (NF)-κB translocation in cells - a phenomenon that leads to exacerbation of acute inflammatory events.

Long-term consumption of diets high in both carbohydrates (including some with traces of lipopolysaccharides) and fat (specifically, saturated fatty acids) such as a high-fat high-carbohydrate (HFHC)-diet by rodents and humans has been reported to cause activation of NF-κB signaling through activation of toll-like receptor 4 (TLR 4), an effect that exacerbates insulin resistance (Shi et al. 2006; Erridge et al. 2007; Anderson et al. 2010; Erridge 2010; Baker et al. 2011). Insulin resistance itself has been shown to be associated with abnormalities such as hyperglycemia and hyperlipidemia that also can also trigger immune dysfunction/over-activation (Robertson et al. 2004; Kolb and Mandrup-Poulsen 2005; Schrauwen 2007; Nikolajczyk et al. 2011; Richard et al. 2017). In turn, prolonged/repeated states of hyperglycemia are known to be a primary factor underlying dysregulation of the innate immune system noted in patients with type 2 diabetes (T2D) (Graves and Kayal 2008).

According to the International Diabetes Federation (International Diabetes Federation (IDF) 2019), T2D accounts for ≈ 90% of all diabetes mellitus cases globally thus making it the most common type of diabetes worldwide. According to IDF statistics, in 2019, there were 19 million diabetics in Africa between ages 20 and 79 years; surprisingly, at the same time, 12 million Africans were reported as living with undiagnosed diabetes. In these populations, the onset of T2D is usually preceded by a state of pre-diabetes that has been reported to last ≈ 20 years (Therrin 2018). Pre-diabetes is deemed an intermediate state between normoglycemia and T2D where blood glucose levels are higher than normal but not yet high enough for a diagnosis of T2D (Rydén et al. 2007). Due to the asymptomatic nature of the condition of pre-diabetes, this has led to challenges in documentation of the prevalence of this condition as well as in physiological changes that occur during this condition (Fonseca 2007; Grundy 2012).

Using an HFHC diet-induced animal pre-diabetes model, previous studies from our laboratory demonstrated there were increases in glycated hemoglobin (HbA1c) levels that were accompanied by increases in host blood pressure, impaired renal handling, and impaired cardiovascular function (Gamede et al. 2018, 2019; Luvuno et al. 2019). These abnormalities were also seen to be associated with an apparent immune activation as demonstrated by changes in host circulating levels of various immune cell types (including those of neutrophils, lymphocytes, basophils, monocytes and eosinophils) during progression of the pre-diabetes state (Mzimela et al. 2019). In addition, there was also an up-regulation of circulating levels of inflam-matory markers, including those of interleukin (IL)-6, tumor necrosis factor (TNF)-α, C-reactive protein (CRP), fibrinogen, P-selectin and soluble cell differentiation 40 ligands (CD40L). While this data from the animal model clearly shows a potential utility in monitoring various immune cell types/inflammatory markers to track pre-diabetes in a host, these findings have not yet been verified in humans with pre-diabetes.

Together, these markers give us insight into whether there is immune activation as the changes in immune cells concentration may be an indicator of glucotoxicity. Additionally, most of these cytokines have been reported to be involved in the physiology of T2D upon activation of inflammatory signaling pathways due to hyperglycemia (Baker et al. 2011; Chawla et al. 2011). The city of Durban (South Africa) is populated by a wide variety of ethnic groups. According to a study by Sosibo and colleagues, this city shows increasing prevalence of pre-diabetes among individuals specifically those 25−45 years of age. Additionally, in this city, have been reported to be affected by different factors such as unhealthy diets and occupational exposures to immunotoxins suggesting a compromise in immune system and exposure to development of neutropenia (Govender et al. 2021). Building on those findings, the present study was undertaken to investigate if – as in the animal models noted above - there were changes in circulating levels of immune cell types and any dysregulation of a select set of inflammatory markers that could corresponded with a presence of pre-diabetes in this age group.

Materials and methods

Chemicals/reagents

All chemicals and reagents used were of analytical grade. The materials and analytic kits utilized here were as follows: Human HbA1c ELISA kit (Elabscience, Houston, Texas, USA); Human Customized Invitrogen “ProcartaPlex”, 4-plex (IL-6, TNFα, sCD40L, and P-selectin) multiplex assay kit (Thermofisher Scientific, Waltham, MA) and, Human CRP and Human Fibrinogen ELISA kits (Elabscience).

Study site, population, and design

The study was carried out at laboratories of the University of KwaZulu Natal (UKZN, Durban, South Africa). A quantitative cross-sectional analytical study was conducted with blood samples (n = 292) collected at King Edward Hospital after UKZN Biomedical Research Ethics Committee (BREC) approval (#BE266/2019). The blood samples were collected from February 2021 to December 2022 from patients of all ethnicities and both genders, who ranged in age from 25−45 years. The selection of samples was done according to selection criteria and data provided by the hospital. The sampling exclusion criteria included: patients <25 and >46 yr-of-age; samples from patients displaying other diseases other than T2D/pre-diabetes; patients with no history of liver disease, thyroid disease, kidney disease, heart disease, depression, HIV; no professional sport athletes; patients under the influence of alcohol and pregnant females. All samples were collected only after a signed informed consent was obtained from each individual.

Pre-diabetes confirmation

To confirm whether samples should be categorized as normal, pre-diabetic, or T2D, the criteria of the American Diabetes Association (American Diabetics Association (ADA) 2016) were applied. Additionally, based on glucose level data obtained from the hospital, HbA1c levels in the samples were measured using a human ELISA kit and following manufacturer instructions. Samples that indicated an HbA1c of <5.7% were considered normal, between 5.7−6.4% pre-diabetic, and >6.4% T2D.

Immune cells and inflammatory markers measurements

An automated Beckman Coulter cell counter (Indianapolis, IN) was used to measure the levels of various immune cell types (e.g. neutrophils, lymphocytes, monocytes, eosinophils, and basophils) in each blood sample. The remaining blood was centrifuged at 3000 rpm for 15 min to obtain plasma that was then collected and stored at −80 °C until used for biochemical analysis.

To measure IL-6, TNFα, sCD40L, and P-selectin in each plasma sample, a customized human Invitrogen “Procarta Plex” 4-plex multiplex assay kit was used, following manufacturer protocols. All results were processed using a Bio-plex MEGAPIX Multiplex reader (BioRad, Hercules, CA). Levels of CRP and fibrinogen in the samples were measured using their respective human ELISA kits. All measurements from the plate wells were obtained using a Spectro star nanoplate spectrophotometer (BMG Labtech, Ortenburg, Baden-Württernberg, Germany). The level of sensitivity of the kits were: 0.23 ng CRP/ml, 5.63 ng fibrinogen/ml, 52.8 ng IL-6/ml, 25.2 ng TNFα/ml, 10.6 ng sCD40L/ml, and 53.900 ng P-selectin/ml. All samples were evaluated in triplicate, following manufacturer protocols.

DATA analysis

All data is expressed as means ± SEM. For the blood levels of the various immune cell types and inflammatory markers (CRP and fibrinogen), data were analyzed using SPSS v.28 software (SPSS, Cary, NC). For these analyses, all groups were compared by applying a one-way analysis of variance (ANOVA) followed by a Tukey-Kramer post-hoc test. For the measures of inflammatory markers obtained with the multiplex assay (e.g. IL-6, TNFα, sCD40L, P-selectin), all data was evaluated using Bio-Plex Manager software v.5.0 and outcomes were then compared using Prism software (v.8; GraphPad, San Diego, CA). For these endpoints, all groups were compared by applying a one-way ANOVA and a Tukey-Kramer post-hoc test. In all cases, a p-value < 0.05 was considered as statistically significant.

Results

The study here utilized a total of 292 blood samples from various test subjects. Based on established parameters, these samples were sub-categorized into three groups (Figure 1): a non-diabetic group (ND, n = 30) with samples from 20 females and 10 males; a pre-diabetes group (PD, n = 90) with samples from 56 females and 34 males; and, a Type 2 diabetes group (T2D, n = 172) with samples from 113 females and 59 males.

Figure 1. Gender distributions within each group. ND: non-diabetes group, PD: pre-diabetes group, T2D: Type 2 diabetes group.

Blood immune cell (neutrophil, lymphocyte, monocyte, eosinophil, and basophil) levels

Neutrophil (PMN) presence in the fresh blood samples was measured in all experimental groups. Results across all three groups indicated that circulating PMN levels were below the expected normal range (NR; 40-60%). The results showed there was a non-significant decrease in PMN levels in the T2D group in comparison within ND subjects (28.4% ND vs 27.0% T2D), p = 0.84 (Figure 2A). The pre-diabetics (PD) also had non-significantly lower blood PMN levels in comparison to the ND hosts (26.7% PD, p = 0.78. The small decrease (0.3%) between the PD and T2D hosts was deemed to fall within the margin of sampling error, p = 0.97.

Figure 2. Comparison of blood immune cells levels in human subjects. (A) neutrophils (PMN), (B) lymphocytes, (C) monocytes, (D) eosinophils, (E) basophils. ND: non-diabetic group, PD: pre-diabetes group, T2D: Type 2 diabetes group. Values shown are means ± SEM.

Analysis of blood lymphocyte levels found that all three groups had circulating lymphocyte levels within the normal range (20-40%). The results showed there was a non-significant decrease in lymphocyte levels in the T2D group in comparison to in the ND hosts (61.6% ND vs. 61.1% T2D), p = 0.99 (Figure 2B). The shift seen in the PD hosts was even less notable (61.2% PD), p = 0.99. The small increase (0.1%) between the PD and T2D hosts was deemed to fall within the margin of sampling error, p = 0.99.

Blood monocyte levels analysis showed that all three groups had circulating levels within the normal range (2-8%). The analysis revealed a non-significant increase in monocyte levels in the blood of the T2D group relative to that seen in ND host samples (4.6% ND vs. 5.4% T2D), p = 0.77 (Figure 2C). Interestingly, the shift in levels was not as great with the PD patients and actually was slightly (albeit non-significantly) reduced (4.4% PD), p = 0.99. Comparisons between the PD and T2D hosts revealed that while the net difference was 1.0%, this difference was not significant, p = 0.46.

Evaluations of blood eosinophil (EOS) levels showed that all three groups had levels within the normal range (1-4%). The results indicated that there was a non-significant increase in blood EOS levels in the T2D group in comparison to in ND subjects (2.41% ND vs. 2.50% T2D), p = 0.98 (Figure 2D). The non-significant increase seen in the blood of PD hosts was even smaller (2.44% PD), p = 0.99. These low levels indicated there was no significant difference in blood EOS levels between the PD and T2D subjects, p = 0.98.

Measures of basophils in the blood showed all three groups had levels within the normal range (0.5-1%). As with the monocyte outcomes, comparisons among the groups showed that vs. both the ND and T2D hosts, there were small non-significant decreases in circulating basophils in the blood of the PD subjects (2.91% ND, 2.44% T2D, 2.35% PD) (Figure 2E). The p values were 0.95 (NT vs T2D), 0.80 (ND vs PD) and 0.85 (PD vs T2D).

Inflammatory markers

Levels of select inflammatory markers (CD40L, P-selectin, IL-6, TNFα) were measured in plasma obtained from hosts in each experimental group. Figure 3A illustrates how there was a non-significant increase in circulating CD40L levels in T2D hosts compared to in the ND group (10.40 pg/ml ND vs. 27.34 pg/ml T2D; p = 0.98). The results showed that while the observed increase in circulating CD40L levels in the blood of the PD group was more substantial relative to levels in the ND group (10.40 pg/ml ND vs. 151.49 pg/ml PD; p = 0.34), this change ultimately was not significant, (nor was the increase relative to the levels seen in the T2D hosts; p = 0.28).

Figure 3. Levels of inflammatory markers in sampled blood. (A) CD40L, (B) P-selectin, (C) IL-6, (D) TNFα. Values shown are means ± SEM. *value significantly different (p < 0.05) between ND and PD, ^#between PD and T2D (p < 0.05).

Figure 3B shows that while there was a non-significant increase in plasma P-selectin levels in the T2D group (442.12 μg/ml) in comparison to in ND hosts (279.23 μg/ml; p = 0.06), there was now a non-significant decrease in plasma P-selectin in the PD group (220.33 μg/ml) relative to that in the ND group (p = 0.70). On the other hand, these depressed PD levels were significantly lower than in the T2D group (p = 0.001).

With respect to plasma IL-6, the data in Figure 3C indicates there was a non-significant increase in circulating IL-6 in the T2D hosts in comparison to in the ND subjects (464.74 ng/ml ND vs.703.64 ng/ml T2D; p = 0.17). In contrast again, compared against levels in the ND group, a non-significant decrease in IL-6 levels was noted in in the PD group (355.51 ng/ml; p = 0.70). Unlike for some of the other markers evaluated here, the net difference between the T2D and PD plasma IL-6 values were significant (p = 0.007).

A different overall trend was noted with respect to plasma TNFα levels (Figure 3D). Specifically, the data revealed there was a non-significant increase in circulating TNFα in the T2D group in comparison to in the ND hosts (25.37 ng/ml ND vs. 82.10 ng/ml T2D; p = 0.49), and that there was an even much greater significant increase in circulating TNFα (150.73 ng/ml) associated with the PD state (p = 0.05). Oddly, the PD levels, albeit almost double that found with the T2D subjects, were not significant different from the T2D levels (p = 0.24).

CRP and fibrinogen

The results in Figure 4A indicate there was a non-significant increase in circulating CRP levels in the T2D group when compared to in the ND hosts (4.27 ng/ml ND vs. 4.57 ng/ml T2D; p = 0.98). In contrast, there was a non-significant decrease in circulating CRP in the PD group (1.93 ng/ml) when compared to the ND hosts (p = 0.31). Oddly again, these PD levels, albeit almost half that in the T2D hosts, were not significant different from the T2D levels (p = 0.11).

Figure 4. Levels of blood (A) CRP and (B) fibrinogen in samples. ND: non-diabetic group; PD: pre-diabetes group; T2D: Type 2 diabetes group. Values shown are means ± SEM.

Analysis of circulating fibrinogen levels (Figure 4B) showed there was a concurrent non-significant increase in the T2D host levels relative to those in the ND subjects (13.40 ng/ml ND vs. 45.63 ng/ml T2D; p = 0.32). Unlike with CRP, in this case, there were concurrent non-significant increases in circulating fibrinogen in the PD group (34.52 ng/ml) compared with in the ND subjects (p = 0.63). Though fibrinogen levels were increased in both groups relative to in the non-diabetics, these levels were found ultimately to not significantly differ from one another, (p = 0.82).

Discussion

One of the complications associated with Type 2 diabetes (T2D) is a dysregulation in the innate immune system (Graves and Kayal 2008). This dysregulation has been reported to be a result of the chronic hyperglycemia observed in T2D subjects (Evans et al. 2002; Monnier et al. 2006; Evans et al. 2002;). However, before the onset of T2D, there is often a long-lasting state of intermediate hyperglycemia known as pre-diabetes (Grundy 2012; Therrin 2018). Several studies conducted in animal models have suggested that the abnormalities observed in T2D begin during pre-diabetes (Luvuno et al. 2018; Gamede et al. 2019; Mabuza et al. 2019; Mzimela et al. 2019).

The city of Durban in South Arica is a culturally diverse urbanized area characterized by increasing levels of consumption of high calorie diets and sedentary lifestyles. These trends have coincided with an increasing incidence of pre-diabetes, specifically among individuals in the 25 − 45 yr-of-age group (Sosibo et al. 2022). Additionally, KZN people, which include Durban, has been reported to be affected by factors that contribute to immune toxicity such as chronic consumption of high calorie diets, sedentary lifestyles and occupational exposure to immune toxifying agents (Govender et al. 2021). To date, there has been no research done to investigate potential changes in immune cell/inflammatory markers associated with either T2D or pre-diabetes in this population. Such information could prove useful for earlier detection of the onset of diabetes, and thus afford an earlier start to treatment or initiation of changes in lifestyle. To gain insight into identification of potential markers of pre-diabetes in this age group, the present study was undertaken to evaluate changes in circulating immune cells as well as in levels of select inflammatory markers in 25−45-yr-old patients with prediabetes in the city of Durban, South Africa. The outcomes would then hopefully build upon results of a previous study that investigated the effects of pre-diabetes on immune cells in an animal model of diet-induced pre-diabetes (Mzimela et al. 2019). For this discussion, the outcomes regarding the neutrophils, lymophocytes and monocytes are addressed; the data showing minimal impact on blood eosinophil and basophil levels allows for those cell types to not be discussed further as potential markers of pre-diabetes in this age group.

While neutrophils (PMN) are needed by the immune system to fight invading pathogens and in injury healing (Honda et al. 2016), the current study detected generalized neutropenia in all the different groups evaluated. One could speculate that this could be due to decreased production or differentiation of PMN in the bone marrow, an event potentially related to effects from overall nutritional disparities in these hosts (Govender et al. 2021). The province of KwaZulu-Natal (KZN) wherein Durban is located, is characterized by an odd co-existence of under- and over-nutrition (Govender et al. 2021). This suggested to us that even though people in this area could still be categorized as non-diabetics, they ultimately can be affected by different factors arising from the local environment, including dietary habits. Other factors that could be contributing to a state of neutropenia are gastrointestinal disorders (which lead to repeated inflammatory states) as well as personal/occupational exposures to immunotoxins. The results presented here showed there seemed to be a trend toward a decrease in circulating PMN levels in T2D hosts. Such results would be in keeping with what is known about PMN during T2D as PMN migrate to chronically inflamed areas such as adipose tissue and endothelial cells (Daryabor et al. 2020), and thus are less present in the blood at any given moment. The findings here appear to extend this trend to the PD state. Still, it is interesting that states of hyperglycemia and hyperlipidemia induce a chronic inflammation condition that is expected to stimulate PMN production, thereby increasing circulating PMN levels (Soehnlein et al. 2017). Clearly, some-thing is occurring during the development of PD and subsequent progression to T2D that allows for circulating levels of these cells to drop even during ongoing states of hyperglycemia and hyperlipidemia.

Lymphocytes are also implicated in the pathology of diabetes (Hampton and Chtanova 2019). For example, studies have shown that in T2D, there are increased levels of circulating activated T-cells due to chronic hyperglycemia and these T-cells secrete cytokines such as IL-6 and TNFα that contribute to the immune dysregulation associated with diabetes (Butcher et al. 2014; Xia et al. 2017). In addition, CD4 and CD8 T-cells will migrate to adipose tissues upon activation and cause local cells to release inflammatory cytokines that further promote the pathology (McLaughlin et al. 2014). In the present study, in the PD group, there was a decrease in plasma lymphocyte levels relative to in ND hosts. The PD results suggest that it is likely that during any induced hyperglycemia/inflammation in pre-diabetes, lymphocytes are recruited to inflamed areas where they then secrete additional inflammatory cytokines like TNFα and IL-6. That these PD hosts also displayed relative increases (vs. in ND hosts) in plasma CD40L suggests to us that the immune cells secreting the CD40L and TNFα were more activated.

It may be that lymphocytes were not the only cell source impacted by the pre-diabetes and that contributed to the observed changes in select cytokine/inflammatory protein express noted. For example, monocytes have also been reported to release TNFα and IL-6 due to hyperglycemia (Chomarat et al. 2000; Nikiforov et al. 2017). Monocytes (which can differentiate into either antigen-presenting dendritic cells or macrophages upon reproduction and activation [Chomarat et al. 2000; Shrestha et al. 2014; Mustafa 2022]) can play an important role in exacerbating T2D as they secrete IL-6 (via induction of protein kinase C (Ngcobo et al. 2022). In T2D, monocytes are recruited to inflamed areas such as adipose tissue (Xu et al. 2015), with the latter sties being a source of monocyte chemoattractant protein-1 (MCP-1) (Degirmenci et al. 2019). In a self-promoting manner then, the newly released MCP-1 induces monocyte migration to the inflamed area. In addition to MCP-1, monocytes are also a good source of IL-6, IL-8, TNFα, and IL-1β, each of which can then act to exacerbate any ongoing inflammation during hyperglycemia as well as during T2D (Jagannathan-Bogdan et al. 2011; Ngcobo et al. 2022). The results in the present study showing a slight increase in circulating monocytes in the T2D but a nominal decrease in the PD group suggested to us that as a moderate hyperglycemia and chronic sub-clinical states of inflammation eventually gave rise to T2D, more and more monocytes were likely being recruited to inflammatory sites. An explanation then for why there was an increase in relative levels of blood monocytes in T2D hosts compared to in the normal hosts remains elusive.

The population of the current study contained more women than men per group. Therefore, it is important to consider the potential impact of sex-related hormones on the results of markers and immune cell obtained herein. One protein whose circulating levels are known to be affected by gender is CRP (Gaskins et al. 2012). This is odd in that CRP is routinely used as a clinical predictor of cardiovascular disease in both males and females. CRP is produced in the liver and its release is induced by circulating IL-6 and TNFα (Lee et al. 2009). Stimulation of adipose tissues has been reported to result in elevated CRP levels in T2D patients. Indeed, the results obtained in the present study showed an increase in CRP levels in the T2D hosts. It would be logical to then surmise that during the chronic hyperglycemia/inflammation that led to T2D, one would expect a triggering of CRP release from the liver. Unexpectedly, in the PD group here, plasma CRP levels were decreased in comparison with those in the ND hosts. One potential explanation for this decrease can be gleaned from the findings of Gaskins et al. (2012) where it was seen that among female subjects, decreases in plasma CRP levels were common due to endogenous estradiol. It could very well be here, that because the PD group was populated primarily with women (see Figure 5), these observed decreases in circulating CRP were artefactual and more heavily impacted upon by the hormone than by any hyperglycemic state in the hosts.

Figure 5. Mean blood concentration of CRP as a function of gender within each group. ND: non-diabetes group; PD: pre-diabetes group; T2D: Type 2 diabetes group.

Conclusions

There remains much to be done to better understand the changes that occur during pre-diabetes that led to the development of T2D. Activation of the immune system and inflammation have each been shown to contribute significantly to this process. The results obtained from the present study suggest that that there are other factors that can contribute to the changes observed such as gender, race, and age. The observed changes in immune cell levels and some of the evaluated inflammatory markers indicate it is increasingly likely that chronic consumption of high-calorie/high-fat diets and living sedentary lifestyles by this Durban population in is having a multiplicity of effects. These seem to include immune system activation and inflammation during the pre-diabetic state that only is amplified as the pathology progresses to T2D.

While these findings provide a basis for more refined marker-defining studies of pre-diabetics, as authors, we wish to acknowledge key limitations in the current study. Of note, a lack of equal numbers of male and female subjects per group was likely a limiting factor in that this imbalance may have skewed some of the results. Further, the current study could not measure all the various inflammatory markers (and hormones) possibly involved in host immune responses due to constraints of sample, time and funding. Follow-on studies will be better designed to over-come these limitations and expand the scope of endpoints measured in the three groups.

Acknowledgments

The authors would like to express gratitude to Mr. Dennis Makhubela for his technical expertise. The authors are also grateful to the King Edward Hospital for the samples used for the study, as well as the National Research Foundation for providing funding (South Africa).

Disclosure statement

No potential conflict of interest was reported by the author(s). The authors alone are responsible for the content of this manuscript.

Data availability statement

The datasets used/analyzed in the current study are available from the corresponding author upon reasonable request.

Additional information

Funding

This work was funded by the National Research Foundation (Grant #106041).