Pediatric Rheumatology Online Journal

Vol 2, No. 2 2004 (119-133)

Leukocyte activation in the HyperImmunoglobulinemia D and periodic fever syndrome

Joost Frenkel¹, Stephanie E. Smetsers¹, Ger T. Rijkers², Jacobus F. Gaiser², Sander M. Houten³, Hans R.Waterham³, and Wietse Kuis²

Affiliations:

Departments of General Pediatrics and
Pediatric Immunology, Wilhelmia Children's Hospital, University Medical Center Utrecht, The Netherlands
Laboratory Genetic Metabolic Diseases, Departments of Pediatrics/Emma Children's Hospital and Clinical Chemistry, Academic Medical Center, University of Amsterdam, The Netherlands

Keywords: Mevalonic acid; fever; hyperimmunoglobulinaemia; immunoglobulin D; autoinflammatory; inflammation; leukocytes; surface antigens

Contact:

Joost Frenkel MD

Div. of Pediatrics Wilhelmina Children's Hospital

KE.04.133.1, University Medical Center Utrecht

PO-box 85090, 3580AB Utrecht, The Netherlands

e-mail: j.frenkel@wkz.azu.nl

tel: +31 30 2504001

fax: +31 30 2505349

Abstract

Objective:

The Hyper IgD and periodic fever syndrome and mevalonic aciduria are characterized by recurrent episodes of generalized inflammation. Both syndromes are caused by a deficiency of mevalonate kinase. How this inborn error in isoprenoid biosynthesis leads to inflammation and which cells are involved in this process is as yet unknown.

We investigated whether specific leukocyte populations are activated during the fever attacks in children with mevalonate kinase deficiency.

Methods:

Blood samples obtained during and between fever attacks were analyzed by white-cell and differential counting and by flow-cytometry. Cells were studied for the expression of CD3, CD4, CD8, CD14, CD20, CD23, CD64, CD69 and HLA-DR.

Results:

Six patients were studied. During fever, monocyte numbers rose 3-fold and neutrophil granulocytes 4-fold. These cells were activated, as reflected by the expression of CD64, which was increased 3-fold on monocytes and 6-fold on granulocytes. There were no such changes in other leukocyte subsets.

Conclusions:

Activation of monocytes and neutrophil granulocytes is involved in the fever attacks of the Hyper-IgD and periodic fever syndrome.

Introduction

The hyperimmunoglobulinemia D and periodic fever syndrome (HIDS), also known as Dutch type periodic fever (MIM#260920), is an autosomal recessive disorder characterized by febrile attacks recurring at more or less regular intervals and the presence of an elevated serum IgD concentration (>100 IU/ml) (1). Over 170 patients have been diagnosed with the disease worldwide (2). Clinical features during the febrile attacks include cervical lymphadenopathy, splenomegaly, hepatomegaly, skin rash, oral ulcers, vomiting, diarrhea, arthralgias and arthritis. Patients often complain of malaise, chills, headache, nausea or abdominal pain (3). During these febrile crises, blood tests reflect an acute inflammatory state with leukocytosis and elevated acute phase reactants e.g. C-reactive protein.

The underlying genetic defect of the syndrome is a deficiency of the enzyme mevalonate kinase (MK) due to mutations in its encoding gene, MVK (4,5) Mutations in the same gene are responsible for mevalonic aciduria (MA, MIM#251170), a syndrome with episodic fever, mental retardation and dysmorphic features (6,7). MK catalyses the phosphorylation of mevalonic acid into 5-phosphomevalonate, an early step in the isoprenoid biosynthesis pathway. This route produces cholesterol which, in turn, is a precursor for steroid hormones, bile acids and vitamin D. Furthermore, this anabolic route yields a number of non-sterol isoprenoids. The latter are hydrophobic molecules such as dolichol, polyisoprene side chains, such as those attached to heme-A and ubiquinone, and the farnesyl and geranylgeranyl side chains of isoprenylated proteins. Currently, the chain of events that links this metabolic defect to episodic inflammation is understood only partly. Indeed, much has been learned about the soluble pro- and anti-inflammatory mediators involved in HIDS. Patient serum contains high levels of pro-inflammatory cytokines such as interferon-γ and interleukin (IL)-6 (8,9) during fever attacks. Ex-vivo, isolated mononuclear cells (MNC) from HIDS patients secrete more IL-1β, IL-6, and TNF-α than MNC from healthy individuals.

It is as yet not known which leukocyte subpopulation(s) are involved in the inflammatory activation that characterizes HIDS. We therefore analyzed leukocyte (sub) populations in peripheral blood of patients during and between fever attacks.

Our aim was to establish whether there were quantitative changes in leukocyte subpopulations and whether one or more of these cell types appeared to be activated in mevalonate kinase deficiency.

Patients and methods

After ethical review board approval, our pediatric mevalonate kinase deficiency patients with either the HIDS or the MA phenotype, were approached to participate in the study. After written informed consent by the patients' parents, blood was drawn by venipuncture in sterile pyrogen-free heparinized plastic tubes as well as in 500 μl EDTA- anticoagulated plastic cups. This was done both when patients were free of symptoms and within the first 24 hours of a febrile attack. EDTA samples were used for white blood cell counting and white cell differential counting (Cell-Dyn 4000, Abbott Diagnostics, Santa Clara, CA). The heparinized samples were kept on ice from the moment of venipuncture until flow cytometric analysis in order to prevent granulocytes from being activated. Patients were not taking anti-inflammatory drugs at the time of blood sampling. Healthy adult volunteers served as controls.

100 μl aliquots of heparinized whole blood were incubated on ice during 30 minutes with diluted antisera. These monoclonal antibodies were directly labeled with Fluorescine isothiocyanate (FITC) or phycoerythrin (PE) and added to a final dilution of 1:10 for anti CD64-FITC (Immunotech, Marseille, France) and 1:100 for all other monoclonals. These were anti-CD3-FITC with anti HLA-DR-PE, anti-CD4-FITC with anti-CD8-PE, anti-CD69-FITC with anti-CD3-PE, anti-CD20-FITC with anti-CD23-PE and anti-CD14-PE (Becton Dickinson Immunochemistry systems, San Jose, CA). Samples were then washed with phosphate buffered saline (PBS) with 1% bovine serum albumin (BSA) and 0.02% NaAzide and after centrifugation at 1500 rpm red cells were lysed during 20 minutes with 2 ml FACS Lysing Solution (Becton Dickinson Immunochemistry systems, San Jose, CA), centrifuged and resuspended twice in 0.01% Na-Azide in PBS. Immunofluorescence was measured with a FACS Calibur flow cytometer and analyzed with Cell Quest software (Becton Dickinson Immunochemistry systems, San Jose, CA). Cell populations were identified by scatter pattern (granulocytes vs. mononuclear cells). Subpopulations with a specific surface marker profile were expressed as percentage of mononuclear cells. Expression of CD23 and CD64 was measured as mean fluorescence intensity (MFI) of those markers on the (sub)population studied. To correct for inter-assay variability of MFI measurements, the MFI was expressed as the ratio of the value obtained in the patient sample and that obtained in a simultaneously tested sample of a healthy adult control.

Statistical analysis:

Cell counts during and between fever attacks were compared by the paired Student t-test, a 2-tailed p value of <0.05 being considered significant. Fluorescence intensity was similarly analyzed. Values are displayed as mean ± the standard deviation. All analyses were performed using GraphPad Prism 3.0 software (GraphPad Software Inc., San Diego, CA)

Results:

Patients:

Six mevalonate kinase deficiency patients participated in the study. Five of these had the HIDS phenotype. These have been described previously as Numbers 119, 132, 135, 137 and 139. One child had the mevalonic aciduria phenotype (10).

In all patients, mevalonate kinase deficiency has been established and mutations have been identified in both alleles of their MVK-genes. Four were compound heterozygous for the mutant alleles 1129 G>A (V377I) and 803 T>C (I268T), and one patient carried the V377I and 59 A>C (H20P) alleles. The mevalonic aciduria patient had only 0.12% residual mevalonate kinase activity due to mutations 1000G>A and 421-422insG in his MVK genes (11).

White blood cell and differential counts were obtained during and between febrile episodes in all 6 patients. In five of these, the data was obtained during 2 fever episodes and in 2 or 3 non-febrile intervals. Immunocytochemical analysis was performed at least once during fever and once between attacks.

White blood cells:

The fever attacks were characterized by leukocytosis which sometimes was extreme (>40x10⁹/l). Leukocytes on average rose from 7.7x10⁹/l between attacks to 18.9x10⁹/l during fever (p=0.001).

Lymphoid cells:

During fever attacks, the percentage of lymphocytes in the differential count decreased 4-fold. This was largely due to a rise in myeloid cells. Absolute lymphocyte numbers showed little decrease (Table 1).

T-lymphocytes:

Absolute T-cell numbers, as determined by CD3 expression, were similar between and during fever attacks. There was a slight decrease of T-cells as a percentage of mononuclear cells during fever (Table 2). However, the absolute T-cell number and the percentage of T-cells were normal for age (12). CD4 and CD8 subsets were also normal for age and remained stable during and between attacks (Table 2) as did the CD4 / CD8 ratio.

The proportion of T-cells expressing CD69, an early activation marker, between attacks (0.4-1.4%) was comparable to that observed by us in 41 healthy adult controls (1.1-1.4%). During fever there was a small increase in CD69 expression (Table 2). The percentage of T-cells that expressed HLA-DR was very similar between and during fever and well within the normal range (13).

B-lymphocytes:

The percentage of B-lymphocytes, as determined by CD20 expression, remained normal between and during fever episodes, as did absolute B-cell numbers (Table 2). B-cell activation was assessed by the MFI of the Fc-epsilon-Receptor II, CD23, on CD20 positive cells. B-cell activation between attacks was comparable to that in healthy adult controls. During fever there was no significant change.

Monocytes:

Monocyte numbers rose 2.8-fold during fever attacks. This was not reflected in the monocyte percentage in the differential count because of a concomitant rise in granulocytes (Table 1). Activation of monocytes was assessed by the presence on the cell surface of CD64 (Fc-gamma Receptor I). Monocytes were identified by the expression of the lipopolysaccharide receptor CD14 (Figure 1a). Activation was quantitatively expressed as the ratio of MFI of CD64 on CD14 positive cells of patients over that on CD14 positive cells of healthy adult controls (Table 2). During fever there was a 2.5-fold rise in CD64 expression on monocytes (p=0.012, Figure 1b).

Granulocytes:

Neutrophil leukocytosis during attacks was striking (Table 1) with absolute neutrophil counts rising over 4-fold (p<0.001). Band forms were present in a minority of patients during fever, but could be as high as 15%. Eosinophil counts decreased during fever (Table 1).

CD64 expression was measured on granulocytes, i.e. on cells that had the light scattering characteristics of granulocytes on flow-cytometry (Figure 1c) and were negative for CD14. The MFI of CD64 rose 6.5-fold during fever (p<0.01).

These changes in both number and degree of activation of monocytes and neutrophil granulocytes constituted the main abnormalities observed in mevalonate kinase deficient patients (Figure 1).

Discussion

Mevalonate kinase deficiency leads to recurrent bouts of generalized inflammation. The chain of events linking the metabolic defect to the inflammatory phenotype is incompletely understood. Somehow, inflammatory effector mechanisms are activated. We aimed to determine in which cell population this occurred. Several leukocyte subsets could be expected to be activated in this disorder. B-lymphocytes might be involved, as suggested by the polyclonal elevation of IgD and IgA, typical of HIDS (14). T-lymphocytes had been implicated previously, because of the high serum concentrations of interferon-gamma during fever attacks (15). Cells of the monocyte/macrophage lineage were expected to be activated. Inflammatory mediators typically produced by such cells have been found to be secreted in increased amounts either in-vivo, ex-vivo or both (16;17). Also, the raised urinary neopterin excetion is indicative of activation of mononuclear phagocytes (18). Finally, neutrophil granulocytes could be involved, since granulocytosis is a well-known feature of the fever attacks in mevalonate kinase deficiency (19).

Despite the small number of patients studied, our data indicate that it is mainly the non-specific immune system that is activated during fever episodes. This finding may very well not be specific for HIDS, but we did not study leukocyte activation in patients with other periodic fever syndromes nor, to our knowledge, did others.

We could not detect signs of B-cell activation. The absolute number of T-cells during fever did not change significantly, but there was a slight decrease in the percentage of T-cells. This was largely due to the increase of monocytes within the mononuclear cell population. There was also no increase in HLA-DR expression. The increase in CD69 expression on T-lymphocytes, though statistically significant, was very modest. Its biological significance, therefore, remains uncertain.

In contrast, there was a 4-fold increase in the number of neutrophil granulocytes and a 3-fold rise in monocyte number. Moreover, these cells were activated as reflected by the raised expression of CD64.

It can not be excluded that these phagocytic myeloid cells are activated indirectly by some other cell population. Blood sampling during attacks took place as soon as fever had become manifest, so any activation preceding the onset of fever would not have been detected. It is conceivable that T-lymphocytes are involved in the initiation of the fever episodes, since these attacks are often triggered by immunizations or infections and the serum concentration of T-cell derived cytokines is elevated in HIDS. Also, the activation of non-circulating cells, such as plasma cells (responsible for IgA ad IgD secretion), sessile macrophages or dendritic cells, would not have been detected by the present study.

However, the analogy with the other hereditary periodic fever syndromes would favor a central role for granulocytes and monocytes. Like HIDS, these are genetically determined autoinflammatory diseases, i.e. disorders in which inflammation is prominent but neither infectious organisms nor auto-reactive lymphocytes or auto-antibodies are involved (20). In two of these, Familial Mediterranean Fever (FMF), the Cold-Induced Auto-Inflammatory syndrome / Muckle-Wells / CINCA syndrome spectrum, the affected gene is expressed exclusively in granulocytes and monocytes (21-24). In the third autoinflammatory disorder for which the gene defect is known, the autosomal dominant TNF-Receptor Associated Periodic Syndrome, the 55kD high-affinity receptor for Tumor Necrosis Factor-alpha is mutated