A 19 year old male presented to an immunodeficiency practice with a history of peri-rectal fistulas at 7 years of age, followed by a deep left neck abscess refractory to antibiotics at 10 years of age. In general, he had a history of at least 1 skin infection per year. The causal microbe was usually methicillin-sensitive Staphylococcus aureus (MSSA) with no evidence of Aspergillus, Nocardia, Pseudomonas or Serratia species. At presentation in the recent visit he reported a peri-rectal abscess one month prior and bloody diarrhea for 1 week with sharp, diffuse abdominal pain, nausea and vomiting, fever, chills and a weight loss of 12 lbs. He was unresponsive to high-dose steroids. His laboratory data revealed both IgA and IgG antibodies to Saccharomyces cerevisiae, no evidence of Clostridium difficile and the stool culture was also negative for any pathogenic organisms but positive for leukocytes. Colonoscopy showed abnormal wall thickening of all segments of the colon and rectum. A diagnosis of severe colitis and perianal fistula was initially provided, and the rectal biopsy revealed moderate colitis with acute cryptitis and focal abscess formation. The childhood history of fistulas and abscesses with Staphylococcus raised concerns for Chronic Granulomatous Disease (CGD).
Laboratory evaluation was performed for neutrophil oxidative burst using dihydrorhodamine (DHR) flow cytometry before and after stimulation of neutrophils with Phorbol Myristate Acetate (PMA) (Figure - normal, healthy donor and 3B - patient). There was no evidence of DHR fluorescence after stimulation in the majority of the neutrophils (96%) consistent with a phenotype observed in X-linked CGD (XL-CGD) (Figure ). However, it was interesting to note that 4% were positive for modest levels of DHR fluorescence after stimulation, which may be suggestive of somatic mosaicism due to spontaneous reversion in a subset of neutrophils. Genetic testing was performed with full-gene sequencing and revealed a nonsense mutation (R130X) in exon 5 of the CYBB gene, which encodes the gp91phox protein (Figure ). This result along with the flow cytometry data was consistent with a diagnosis of XL-CGD. Flow cytometric analysis (Figure ) and genetic testing (data not shown) was performed on the mother of the patient and revealed that she was not a carrier of the disease-causing mutation, and therefore, the patient had a de novo or spontaneous mutation that accounted for his clinical phenotype of CGD.
Figure 3 Evaluation for Chronic Granulomatous Disease (CGD). A) Flow cytometric analysis for neutrophil oxidative burst (NOXB) in a healthy control. B) Flow cytometric analysis for neutrophil oxidative burst (NOXB) in a patient with X-linked Chronic Granulomatous (more ...)
A second patient, a 23 year-old female was seen in the same immunodeficiency clinic as the above-mentioned male patient. The female patient was diagnosed with Crohn's disease at the age of 13 years when she had abdominal pain, fatigue and hematochezia. She underwent exploratory endoscopy and colonoscopy and her biopsy showed evidence of mild to active small bowel and colonic colitis with non-necrotizing granulomas. Her prior history was significant for skin abscesses, at least once per year, on the upper arm, gluteal region, thighs, vulvar and vaginal areas. There was no evidence of pneumonia, sinusitis, osteomyelitis, cellulitis or meningitis. She was treated almost continuously with immunosuppressive and biological therapies along with steroids since the initial diagnosis of Crohn's disease. Her family history was remarkable for XL-CGD and ocular complications of CGD. Flow cytometric testing for neutrophil oxidative burst revealed 2 populations for DHR fluorescence with a larger negative and smaller positive population (Figure ). Genetic testing revealed a heterozygous deletion of 16 nucleotides (c.360-375del16). The patient's mother and two maternal aunts carried the same deletion mutation (one of these maternal aunts also had ulcerative colitis and primary biliary cirrhosis), and one maternal uncle died at the age of 18 months with recurrent neck abscesses. The family history also revealed two maternal great-uncles who died in childhood of unknown causes, but presumed CGD.
The clinical history of inflammatory bowel disease (IBD), recurrent skin abscesses (facial, labial, peri-rectal), poor surgical wound healing, aphthous ulcers and ocular complications all suggest a clinical phenotype of XL-CGD, due to skewing of X-chromosome inactivation (lyonization). The DHR flow cytometry results indicate that there at least 30% neutrophils with normal oxidative burst function. Similar analyses done elsewhere showed positive DHR populations between 19-26%. It has been reported that if there are greater than 10% of neutrophils with normal oxidative burst, there is typically no evidence of a clinical phenotype [47
CGD is a relatively rare primary immunodeficiency with an incidence of approximately 1 in 200,000 to 250,000 individuals characterized by defects in the oxidative burst pathway that is linked with phagocytosis in myeloid cells, such as neutrophils. The primary defect in CGD is associated with the key enzyme involved in generation of the respiratory burst, NADPH oxidase. This enzyme has at least 5 subunits (Figure ), two of which are membrane-bound, gp91phox (CYBB
gene) and gp22phox (CYBA
gene), and three are cytosolic components, p47phox (NCF1
gene), p67phox (NCF2
gene) and p40phox (NCF4
gene). The p40phox primarily interacts with p67phox and forms a larger complex with p47phox, which in turn interacts with a RacGTPase, RAC1
, permitting translocation to the membrane upon stimulation where it activates the catalytic core of the NADPH oxidase formed by the gp91phox and p22phox proteins. The most common form of CGD is X-linked accounting for approximately 70% of cases, due to mutations in the CYBB
gene. The remaining 30% of cases are associated with mutations in the other subunits and inherited in an autosomal recessive (AR) manner. Mutations in NCF1
account for ~25% of the AR cases, while NCF2
mutations are quite rare. The most recent NADPH subunit in which mutations were found to be associated with CGD was the p40phox (NCF4
) reported in a single patient [51
Clinically, CGD is characterized by recurrent bacterial and fungal infections of primarily the lungs, gastrointestinal tract, skin, and lymph nodes [52
] caused largely by a relatively small number of pathogens - Staphylococcus aureus
species, Serratia marcescens
. Most of these pathogens are catalase-positive organisms. The most common clinical manifestations are pneumonia, cutaneous abscesses, lymphadenitis and chronic inflammatory reactions resulting in granulomas.
Carriers of XL-CGD and AR-CGD are usually asymptomatic, however, about 50% of XL-carriers have been reported to have recurrent mouth lesions, manifesting as either gingivitis or stomatitis. Further, skewing of X-chromosome inactivation (lyonization) with inactivation of the normal X-chromosome has been reported in CGD, which could potentially confer a mild clinical phenotype in the female carrier, though this typically does not happen until the proportion of skewed, inactivated neutrophils drops below 10%, as stated previously, [47
], though healthy carriers with less than 10% normal neutrophils have also been reported [53
]. The female carrier for XL-CGD presented in this article had, at all the time-points tested, greater than 10% neutrophils that were positive for oxidative burst, yet there was evidence of a clinical phenotype with recurrent skin infections and the IBD-like colitis. Further, age-related changes in X-chromosome inactivation patterns have been shown to change the relative proportion of normal to abnormal neutrophils conferring a clinical phenotype on female carriers as they age [46
Laboratory diagnosis of CGD can be achieved by performing flow cytometric analysis to evaluate NADPH oxidase activity (oxidative burst) using dihydrorhodamine (DHR) 1,2,3 as a fluorescent marker of hydrogen peroxide generation. This is a relatively rapid and highly sensitive assay and allows the use of whole blood without purification of neutrophils, and is reasonably stable allowing measurements to be performed up to 48 hours after blood collection. Due to these reasons, this assay has replaced superoxide measurements and the Nitroblue tetrazolium (NBT) slide test as the primary screening assay for CGD [46
Genetic testing is used for identification of the specific gene (encoding a subunit of NADPH oxidase) and relevant mutation. For the majority of CGD cases, gene sequencing of the CYBB
gene permits identification of the causal mutation. The majority of mutations (~70%) in this gene are single nucleotide changes, which include splice-site, nonsense and missense mutations, while the remaining ~30% of mutations are deletions and/or insertions [57
DHR-based flow cytometry can also be used to identify patients with AR-CGD (Figure ), though this can be trickier to interpret and requires a certain level of skill as well as a more quantitative reporting format, which includes both the frequency of neutrophils positive for oxidative burst after PMA stimulation and the intensity of fluorescence per cell (MFI) [55
]. Since there are 4 genetic defects (CYBA
) associated with AR-CGD, one would either have to do mutation analysis for all four genes, which could be cost-prohibitive, or do additional second-tier screening tests, such as intracellular flow cytometry for the various subunits - p22phox, p47phox and p67phox [58
] or immunoblot analysis prior to genetic testing. These are not widely available in clinical labs and are probably most often done in the research setting, which may, by default, necessitate genetic testing to identify the specific gene defect.
Flow cytometry can also be used for carrier detection for XL-CGD, which should typically reveal a mosaic pattern for DHR fluorescence. However, it should be kept in mind that the nature of random X-chromosome inactivation could result in either a near-normal or a highly abnormal pattern in the flow analysis for oxidative burst in female carriers. Therefore, genetic testing remains the most robust way to perform carrier identification, especially if the familial disease-causing mutation is known. The flow-based DHR test is not sensitive enough to identify obligate carriers (parents of patients) or sibling carriers of AR-CGD caused by NCF1
or NCF 2
mutations as there appears to be normal oxidative burst on stimulation of neutrophils (Figure ), and the assay has not been tested for CYBA
carriers. Therefore, detection of AR-CGD carriers is best performed by genetic testing, though this can pose challenges with regard to the NCF1
gene, since several unrelated patients have been reported to have a dinucleotide deletion (ΔGT) in exon 2 of this gene [59
]. A recombination event between the functional NCF1
gene and two pseudogenes, on the same chromosome, carrying this ΔGT leads to the incorporation of the deletion into the NCF1
gene. This phenomenon renders carrier testing for p47phox defects difficult because normal individuals are apparently heterozygous for this GT deletion due to the pseudogenes. There are potential solutions to this problem [63
], and while normals can be distinguished from patients and carriers, it remains unknown whether the "hybrid' protein expressing part of the sequence from the NCF1
gene with part of the sequence from the pseudogenes is really functional [65
], and therefore, only NCF1
-defective patients have been identified so far.
Prenatal diagnosis for CGD can be performed by fetal DNA testing along with gender analysis, if the familial mutation is known, from a chorionic villus sample (CVS) or amniotic fluid cells. The gene sequence from the fetus should be compared to the mother and a symptomatic family member as well as a normal individual to determine to confirm and validate the result. A combination of flow cytometric DHR analysis, genetic testing and family history was useful and relevant in the diagnosis of these two patients with CGD.
As the above cases exemplify, the diagnostic approach for most primary immunodeficiencies include a variety of laboratory tests and techniques, and several, but not all, of these analyses (Table ) can be performed by multicolor and/or multiparametric flow cytometry [2
]. In the case of monogenic defects, genetic testing remains the most valuable test for confirming a diagnosis, providing specific gene and mutation information as well as enabling genotype-phenotype correlations [5
]. The organization and characterization of mutations for specific PID-related genes has become streamlined and widely available through the primary immunodeficiency databases [66
] enabling correlation of new and previously identified mutations with clinical and immunological phenotype, besides family information.
Non-disease-specific immunological tests used for the diagnosis of PIDs
While the above examples showcase the utility of flow cytometry to evaluate specific protein defects in the diagnosis of PIDs, it is also a very versatile tool for immunophenotyping of lymphocyte subsets and assessing lymphocyte or other leukocyte subset functions in PIDs. For example, defects in circulating B cells have been recognized in the very heterogeneous PID -Common Variable Immunodeficiency (CVID) for a number of years, and over time, several classifications involving B cell subsets and immunophenotyping have evolved in an effort to organize and stratify this complex and multifaceted immunodeficiency [11
]. Similarly, T cell immunophenotyping has been used to identify abnormalities or changes in naïve, memory, effector, activated, TH17 inflammatory T cells, regulatory T cells (CD4+CD25+FOXP3+) and recent thymic emigrant (RTE) populations for diagnosis of several combined or cellular immunodeficiencies such as severe combined immunodeficiency (SCID), Omenn syndrome, Hyper IgE syndrome (HIES), IPEX (immunodeficiency, polyendocrinopathy, enteropathy, X-linked), CVID and DiGeorge (chromosome 22q11.2 deletion) syndrome among others [74
Heterogeneity in lymphocyte subsets is not restricted to only T and B cells, but also present in the NK cell compartment, and multicolor flow cytometry can be used to immunophenotype human NK cells in various PIDs where NK cell defects are either primary or secondary [91
]. However, when performing immunophenotyping for circulating lymphocyte subsets, it must be kept in mind that to obtain analytically stringent data, various factors, such as diurnal changes, acute exercise, hormonal alterations, age and gender influence these populations, quantitatively and qualitatively (especially relevant for serial monitoring), and this must be taken into consideration [96
Diagnosis of PIDs with T cell defects also often involves the use of molecular techniques, besides flow cytometry, and these include analysis of thymic function and T cell receptor repertoire diversity [101
]. Quantitation of T cell receptor excision circles (TREC), which are episomal by-products of T cell receptor rearrangement, by polymerase chain reaction (PCR) methods, especially real-time PCR, has been used to determine thymic output [102
]. However, it should be kept in mind that TREC levels are affected by cellular division as well as the longevity of naïve T cells in the periphery [102
] and therefore, may not be always useful as a marker for recent
thymic emigration. But, use of TREC in conjunction with quantitative analysis of naïve T cells and/or recent thymic emigrants (RTE) by flow cytometry [83
]is likely to provide a comprehensive assessment of thymic function. Accurate interpretation of TREC and RTE data requires correlation with total T cell counts along with the use of age-appropriate reference values derived from healthy donors, both pediatric and adults (Hoeltzle et al, manuscript in preparation). T cell receptor (TCR) repertoire diversity can be assessed by flow cytometry, however since the panel of reagents available covers only 2/3rd
of the known TCR -beta gene -variable region (TCR Vβ) families, molecular techniques, such as immunoscope analysis (spectratyping), have been found to be more sensitive and stringent [105
Besides identifying quantitative anomalies in various immune cell populations by flow cytometry, functional assessment of these cell populations is equally important and can be achieved, for the most part, by the same methodology, though other methods can also be used. For example, measurement of lymphocyte proliferation to mitogens, such as Phytohemagglutinin (PHA), Pokeweed mitogen (PWM) and Concanavalin A (Con A), and antigens, such as Candida albicans
(CA) and Tetanus toxoid (TT) to ascertain T cell immune competence in PIDs [110
] has long been performed by DNA incorporation of radiolabeled thymidine (3
H-T) after stimulation of peripheral blood mononuclear cells (PBMCs) with the appropriate agent. Elimination of techniques involving radioactivity is always beneficial to the clinical laboratory, and flow cytometry-based methods, primarily using the intracellular fluorescent dye, CFSE (carboxyfluorescein diacetate succinimidyl ester), are now available for measuring cellular proliferation [111
]. However, a recent study seems to suggest that the use of CFSE to measure lymphocyte proliferation for the diagnosis of cellular PIDs would be inaccurate due to the high rate of false positive results [114
]. CFSE is also difficult to use in a high-throughput clinical laboratory due its light-sensitive nature and the requirement for pre-labeling of cells.
A more attractive alternative has been the direct incorporation into DNA of a non-radioactive compound, an alkyne-modified nucleoside (EdU, 5-ethynyl-2'deoxyuridine), which is fluorescently tagged through covalent interaction with a dye-labeled azide, and used to visualize cell proliferation by flow cytometry [115
]; Erickson et al, manuscript in preparation). The flow cytometry method of measuring proliferation offers several distinct advantages over the radioactive method, besides the obvious elimination of radioactivity, including, the ability to measure cellular proliferation in distinct lymphocyte subsets, and assess cellular viability, apoptosis and death using appropriate markers, such as Annexin V and 7-AAD, in the same assay. Flow cytometry also allows measurement of other cellular functions, such as phosphorylation of proteins involved in cell signaling pathways [117
], though these assays are typically available at present only in larger clinical reference or research laboratories. An example of protein phosphorylation key to immune regulation includes the JAK-STAT pathway [119
], and mutations in at least three STAT family members (STAT1, STAT3, STAT5B) are known to be associated with distinct PIDs [121
Laboratory evaluation is essential not only for the diagnosis of PIDs, but also for the evaluation and measurement of recovery of immune function after therapeutic intervention, especially, but not exclusively, in hematopoietic stem cell transplantation (HSCT) [127
]. However, timely treatment requires early diagnosis, especially of PIDs that are fatal, if left untreated, such as SCID or severe T cell lymphopenia [134
]. The adoption of newborn screening (NBS) for SCID and other T cell deficiencies as part of the NBS panel, by the federal advisory committee on heritable disorders in newborns and children, in 2010 has ushered in a new era of population-based screening for these critical PIDs. The screening protocol involves detection of TREC in dried blood spots, followed by additional confirmatory flow cytometry and genetic testing when appropriate [137
]. Early identification of SCID and T cell-deficient patients through the NBS program will pave the way for these infants to receive rapid intervention resulting in improved overall survival.
In conclusion, laboratory-based testing for PIDs is a rapidly expanding, constantly evolving field that plays an integral role in the diagnostic work-up of these complex immunodeficiencies, but also simultaneously provides valuable insights into human immunobiology. However, quality control and standardization of techniques, methods, platforms and reference values is essential to successful and accurate outcomes for immunological analyses within the laboratory, and clinical trial models may provide a frame of reference for such endeavors [145