Monocyte subsets and their activation markers in Kawasaki disease – A Study from a Tertiary care Center in North India

Background: Kawasaki disease (KD) is a common childhood vasculitis and autopsy studies have shown monocyte inltration in affected blood vessels. Role of monocyte subsets and their activation status in KD have not been completely explored. Materials and methods: Sixteen children diagnosed with KD were included in our study along with 16 age and sex-matched controls. We studied subpopulations of monocytes (classical, intermediate and non-classical), early and late monocyte activation markers during acute and convalescent stages. We also assessed soluble markers of monocyte activation (sCD14, sCD163, CCL2). Results: Signicant elevation in absolute number of classical [CD14+CD16-] and intermediate monocytes [CD14++CD16+] were noted in KD and febrile controls in comparison to healthy controls. CD69 expression in classical, intermediate and non-classical monocytes were higher in KD when compared to healthy controls. Absolute counts of classical and intermediate monocytes expressing CD69 and HLA-DR were signicantly lower in the convalescent stage as compared to acute KD. We observed no signicant elevation in sCD14 and sCD163 levels in KD as compared to controls. However, median CCL2 levels were highest in children with KD. Conclusion: Our results suggest that both early and late monocyte activation occur in acute phase of KD that subsides in convalescent phase.


Introduction
Kawasaki disease (KD) is a common childhood medium vessel vasculitis predominantly affecting coronary arteries. [1] KD is now considered the most common cause of acquired heart disease in children in developed countries. Around 20-25% of untreated children develop coronary artery abnormalities (CAAs) in KD. The disease has gained more attention in recent times due to the overlapping features with multisystem in ammatory syndrome in children (MIS-C) temporally associated with severe acute respiratory syndrome corona virus-2 (SARS-CoV-2). [2] Children of Asian ethnicity are more commonly affected than Africans, Hispanics, and Caucasians. Etiology of KD is currently unknown. KD is likely triggered by an infectious or environmental agent that initiates an abnormal immune response in a genetically susceptible host. [3] Activated monocytes play an important role in the immunopathogenesis of KD by production of proin ammatory cytokines such as TNF-α (tumor necrosis factor), IL1β (interleukin), and IL-6. In the acute stage of KD, increase in circulating CD14+ monocyte counts have been described. Recent studies have documented that CD14+CD16++ monocytes (non-classical monocyte subpopulation) play an important role in in ammation in KD [4]. CD14 along with toll-like receptor 4 (TLR4) functions as a coreceptor for binding to the bacterial lipopolysaccharides (LPS). Several studies have also noted increased levels of circulating soluble monocyte activation markers like sCD14, sCD163, MCP-1 and CCL-2 in acute stages of KD. [5][6][7][8] Although individual markers of monocytes activation have been studied in children with KD, combination of different markers of monocyte activation has not been assessed yet. There is also paucity of literature on different monocyte subpopulations in KD.
In this study, we studied peripheral blood levels of monocyte activation markers (including early and late markers) and different populations of monocytes (classical [CD14+CD16-], intermediate [CD14++CD16+], and non-classical [CD14+CD16++] monocytes) in acute KD, and also in follow-up to assess the impact of intravenous immunoglobulin (IVIg) therapy on monocyte activation.

Results
Sixteen (n=16) children diagnosed with KD were included in the present study. Sixteen children were taken as febrile controls and 16 as healthy controls. The median age group [interquartile range (IQR)] of children with KD was 1.95 years (0.6-4.25) as compared to 2 (1.25-5.5) and 3 (1.25-5.5) years in febrile controls and healthy controls respectively. Male to female ratio of cases, febrile controls, and healthy controls was 2.2:1, 1.6:1, and 0.7:1 respectively. All patients with KD had a fever duration of more than 5 days. Median fever duration in children with KD and febrile controls were 9 (7-15) and 7.5 (6.5-18) days, respectively ( Table 1). Causes of fever in febrile controls included pneumonia in 3, liver abscess in 3, tuberculosis in 3, empyema in 1, enteric fever in 1, urinary tract infection in 1, meningitis in 1, cellulitis in 1, osteomyelitis in 1, and dengue in 1. Among cases with KD, 7 (43%) had incomplete KD and 9 (56%) had complete KD. None had features of shock or multisystem in ammatory syndrome temporally associated with severe acute respiratory syndrome corona virus-2 (SARS CoV-2) infection. Three patients had evidence of CAAs -2 had persistent and 1 had transient CAA (Suppl . Table). All children with KD received IVIg 2 g/kg, 3 children received In iximab, 2 received cyclosporin, 1 received glucocorticoids.   (Figure 2).
Flow cytometric parameters were also compared between acute KD, post-IVIg, and 3rd month follow-up samples in KD cases. Absolute counts of classical monocytes and intermediate monocytes were higher in acute KD and it progressively decreased in post-IVIg and 3rd month follow-up samples ( Figure 3). The absolute counts of CD14+ monocytes expressing CD69 and absolute counts of CD14+ monocytes expressing HLA-DR were signi cantly higher in acute KD in comparison to 3rd month follow-up ( Figure 3).

Comparison Of Monocyte Activation Markers (By Elisa)
ELISA was done for 16 patients with KD to assess soluble activation markers of monocytes (sCD14, sCD163, CCL2). While median sCD14 and sCD163 levels were higher in febrile controls compared to children with KD and healthy controls, median CCL2 levels were highest in children with KD. (Table 3). We observed no signi cant uctuations in levels of sCD14 and sCD163 over time in children with KD. However, levels of CCL2 signi cantly dropped after IVIg infusion in children with KD (Supplementary Table 1).

Discussion
Increased numbers of peripheral blood CD14+ monocytes/macrophages and activated CD14+/CD23+ monocytes/macrophages are seen during the acute stage of KD. They secrete high levels of TNFα and IL-1 in acute KD. In our study, we documented increase in CD14+ monocytes, classical (CD14+CD16-) and intermediate (CD14++CD16+) monocytes during acute phase of KD that normalised in follow-up. Both early (CD69) and late (HLA-DR) activation markers in monocytes were also elevated in acute phase of KD that reduced in follow-up. There were signi cantly elevated levels of CCL2 in acute KD, as compared with post-IVIg and convalescent phase. We observed higher levels of CCL2 in patients with KD and febrile controls as compared to healthy controls.
In our cohort, we found signi cantly increased AMC in acute stages of KD as compared to healthy and febrile controls. Majority of the previous studies in KD have focussed on mononuclear cell counts rather than just monocyte counts. Matsubara et al. found no signi cant difference in mononuclear cell count between KD patients and control but monocyte count (CD14+) was signi cantly higher in patients with acute KD as compared to control subjects [10]. Several other studies have documented elevated levels of different serum cytokines (TNF-alpha, IL-1, and IL-6) that are reckoned to be produced by monocytes in the acute stage of KD [11] Matsuguma C et al studied the difference in absolute monocyte counts between IVIg responsive and IVIg resistant groups. They found AMC to be signi cantly higher in IVIg responsive group as compared to IVIg resistant group [12].
Classical monocytes [CD14+ CD16-], which constitute the majority of monocytes, act as scavengers toward apoptotic cells and assist in resolving in ammation [13].  [17]. To the best of our knowledge, no study has studied nonclassical monocytes [CD14+CD16++] in KD and this area requires more research.

CD69 is an early activation marker that is expressed on all hematopoietic cells except RBCs. Crosslinking of this molecule leads to calcium in ux and formation of both cyclooxygenase and lipoxygenase metabolites [18]. Many studies have been done on CD69 expression on T-lymphocytes in KD.
However, studies on CD69 expression on monocytes in KD are lacking. CD69 expression on monocytes has been studied in various other diseases like Alzheimer's disease and sarcoidosis. In our cohort of KD patients, we found an increased percentage expression of CD69 in total monocytes (CD14+) in KD cases and febrile controls as compared to healthy controls suggesting the role of early monocyte activation in KD. HLA-DR is a late monocyte activation marker and is mainly seen on antigen presenting cells (APCs). HLA-DR is a surrogate marker for immune system stimulation and the expression of HLA-DR on monocytes has been shown to determine the e cacy of antigen presentation to T-helper cells [19]. Increased numbers of monocytes with high expression of HLA-DR have been shown in other in ammatory disorders like rheumatoid arthritis and in ammatory bowel disease [20][21].
On comparing acute KD with post-IVIg and convalescent stage we found a signi cantly higher percentage of total monocytes (CD14+) expressing the late activation marker (HLA-DR) in acute stage and post-IVIg as compared to convalescent stage. These ndings indicate increased antigen presentation by monocytes in acute stage. One of the well-known mechanisms of IVIg is to decrease antigen presentation by downregulating MHC-II and dendritic cells and also to decrease T-lymphocyte expansion and to increase apoptosis of T-lymphocytes. Hence, decrease in HLA-DR expression could be related to immunomodulatory action of IVIg [22].
Several studies have documented changes in the levels of different activation markers of monocytes like sCD14, sCD163, CCL2/MCP-1 etc in patients with KD (Table 4). [5][6][7][8] In the present study, we documented signi cantly elevated levels of CCL2 in acute KD, as compared with post-IVIg and convalescent phase, and MCP-1/CCL2 levels were higher in patients with KD and febrile controls in comparison to healthy controls. These ndings were consistent with the results of Asano et al. and Terai et al who observed signi cantly elevated circulating levels of MCP-1 in the acute phase of KD that drastically reduced after IVIg therapy. [7,23] In our study, there was no signi cant rise in sCD14 and sCD163 levels in children with acute KD when compared with febrile and healthy controls. This is in contrast to the results obtained by Takeshita  Azuma et al also noted signi cant elevation in levels of sCD163 following IVIg therapy in initial IVIg responder group, however, the same observation was not noted in IVIg non-responder group. Authors speculated that IVIg could possibly detach CD163 from the surface of macrophages due to cross-linking of Fc-gamma receptors.
[6] We also observed increase in sCD163 levels following IVIg therapy. However, the difference was not statistically signi cant. Small sample size of our cohort could have possibly resulted in not achieving the statistical signi cance. Cheung et al. (2005) showed signi cant overexpression of CCL2/MCP1 genes in THP1 macrophages in children with CAAs who had past history of KD [24]. But this study was carried out in children with KD at mean 7 years after acute illness. These ndings suggest that MCP-1 expression persisted to remain elevated in CAAs in KD.
Our study was unique as multiple monocyte activation markers were assayed in study population using methods like ow cytometry and ELISA. Another strength of this study was inclusion of different subsets of monocytes (classical, intermediate and non-classical) in analysis. Limitation of our study include a small sample size. However, this is understandable given that our study has to be carried out in a limited time span. Studying monocyte activation markers in a large multicentric cohort of patients would provide more convincing data on monocyte activation status in children with KD.
In conclusion, our study showed that both classical and intermediate monocytes along with early and late monocyte activation markers were elevated in acute phase of KD that subsided in follow-up. We also documented elevated levels of CCL2/MCP1 in acute stage of KD implying monocyte activation in the acute phase of KD. Our results suggest IVIg may decrease monocyte activation in KD, thereby controlling systemic in ammation. However, results of our study must bereplicated in larger cohorts with inclusion of more severe forms of KD to assess whether thedegree of monocyte activation could predict IVIg resistance or formation of CAAs.

Study Population
We enrolled sixteen (n=16) children diagnosed as KD in this prospective study from Pediatric Allergy Immunology Unit, Advanced Pediatrics Centre, PGIMER, Chandigarh, India (January 2019 to December 2020

Laboratory Blood Tests
We estimated expression of CD16, CD69, and HLA-DR on CD14+ monocytes. Under sterile aseptic precautions, 2 ml of peripheral venous blood was drawn by venepuncture into the Ethylene diamine tetraacetic acid (EDTA) vacutainer. Samples were collected at three different stages of illness in patients with KD-prior to administration of IVIg and aspirin, 24 hours after receiving IVIg, and 3 months after illness onset (convalescent stage). EDTA samples were processed immediately for ow cytometry assay. One hundred (100) µl of whole blood from the EDTA vacutainer was transferred in two falcon tubes. One tube was labeled as unstained (only blood) and other tube was immunolabelled with the relevant antibodies-CD45, CD14, CD16, CD69, and HLA-DR. The sample with labeled antibodies was then incubated in dark at room temperature for 30 minutes. After incubation, the erythrocytes were lysed twice ( Figure 1) We also measured plasma sCD14, sCD163, and CCL2 levels by ELISA. Plasma was extracted from the EDTA samples by centrifugation (3000 rpm for 5 minutes) and was stored at -80°C refrigerator for future analysis. We used commercially available Quantikine® ELISA kits (R&D Systems® USA) in a fully automated ELISA machine (In nite200 pro) and the results were read in the ELISA reader (Tecan).  Table   Table 4 is available in the Supplemental Files section. Figure 1 Gating Strategy of our study Trend of Absolute counts of CD14+ monocytes expressing HLA-DR in KD patients with CAA as compared to patients without CAAs

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.