C. albicans (ATCC 14053) was grown in liquid media brain heart infusion, BHI (Oxoid - Thermo Fisher, Waltham, MA, U.S.A.), harvested and stored in frozen aliquots at −70°C. For the NADPH oxidase experiments, C. albicans was heat killed (Hk) by incubating an RPMI-yeast suspension (26 × 10⁶ yeast cells) in a water bath at 65°C for 60 min; the failure of fungal growth on Sabouraud agar was used to validate the lack of viability after the heat killing. After performing the preliminary titration experiments, 6 × 10⁶ Hk C. albicans yeast were placed in microfuge tubes, opsonized with 0, 1.5, or 15 mg/mL of human polyvalent IgG (IV IgG, Sandoglobulin CSL Behring, PA, USA) or 3 or 30% of human immune serum (HIS, which was obtained from a patient with recurrent empyema with Candida and Klebsiella infections) for 30 min at 37°C, and immediately transferred to an ice bath. For the killing experiments, live C. albicans were opsonized by incubating 6.5 × 10⁶ yeast cells with IV IgG at 15 mg/mL or 30% of HIS for 30 min at 37°C and 5% CO₂.

The NADPH oxidase activity in blood leukocytes from 11 healthy adult (6 women and 5 men) volunteers aged 26–45 years (median 33 years), who provided informed written consent, was assessed by flow cytometry after oxidation of dihydrorhodamine (DHR) 123 to rhodamine, which was performed using previously described methodology with some modifications (18). Briefly, the blood leukocytes were incubated with Hk C. albicans as follows: for each condition tested, 120 μL of venous blood were mixed with 4 mL of cold lysis buffer (NH₄ Cl 155 mM, NaHCO₃ 14 mM, and EDTA 0.13 mM) in 15-mL conical polypropylene tubes and incubated for 20 min in an ice bath. After centrifugation for 10 min at 400 × g at 4°C, the supernatants were discarded. The leukocyte pellets at the bottom of the centrifuge tubes were gently resuspended, washed again with cold PBS, and finally resuspended in 100 μL of plain RPMI media. Each tube was gently mixed with a micropipette with 50 μL (3 × 10⁵ yeast) of Hk C. albicans (with or without opsonization) and incubated at 37°C and 5% CO₂ for 1 h. A positive methodologic control for the NADPH oxidase activation of leukocytes was included in each experiment by including specimens stimulated with 70 ng of PMA (Sigma Aldrich, St. Louis, MO, U.S.A.) for 30 min. Then, without washing, 1 μg of DHR 123 (stock dissolved in DMSO and then diluted in PBS) (Sigma Aldrich) was added to each tube, followed by an additional 30 min of incubation at 37°C. The tubes were washed with 5 mL of cold PBS and centrifuged at 400 × g for 7 min; then, the supernatants were decanted, and the leukocyte pellets were resuspended in 500 μL of cold PBS and transferred to flow cytometry tubes. The cells were harvested and analyzed using a FACSCalibur flow cytometer (Becton Dickinson, San Jose, CA, U.S.A.). In total, 10⁷ events were acquired and stored per tube and condition, and the results were analyzed by the mean fluorescence intensity (MFI) of rhodamine gating on granulocytes. The stimulation indices per donor and condition were calculated by dividing the rhodamine MFI with opsonized Candida by rhodamine MFI with unopsonized Candida. The statistical analyses were performed using the Wilcoxon test with Prism software (GraphPad software, La Jolla, CA, U.S.A.) to compare the indices of NADPH oxidase activity measured under different opsonization conditions. Statistical significance was considered at p < 0.05.

Subsequently, we evaluated the capacity of blood leukocytes to kill ounopsonized and opsonized C. albicans in experiments using whole blood from healthy adult volunteers based on a previously published method (19) with some modifications. Briefly, heparinized blood samples were placed on a rocker shaker to maintain the blood motility and preserve the cell viability. In total, 300 μL of whole blood were placed in 2-mL plastic vials, inoculated with 300 μL of C. albicans suspension (3 × 10³ C. albicans yeasts with or without opsonization) and incubated at 37°C and 5% CO₂ for 1 h under gentle shaking in a rocker shaker. The infected blood samples were centrifuged at 11,750 × g for 10 min in a microcentrifuge, and the supernatants were discarded. The cells were lysed by adding 1 mL of distilled sterile water and vigorously shaken for 1 min, followed by centrifugation at 11,750 × g for 10 min, and the supernatant was discarded. The Candida pellets were resuspended in liquid media BHI, 10 μL of each serial dilution (10⁻¹-10⁻⁵) were seeded in triplicate on Sabouraud agar, and the CFU was determined after a 24-h after incubation at 37°C and 5% CO₂. The killing of C. albicans was estimated on the basis of the enumeration of the CFUs. The statistical analysis was performed by the Wilcoxon test with Prism software.

Patient 1 (P1), who was an 8-year-old female, was born as the third child of a Mexican mestizo family with a past history of an older sister who died before the age of 5 due to Mycobacterium and Candida infections after suffering from disseminated BCG disease from a BCG vaccine (20). The mutation causing the immunodeficiency in the offspring of the family was traced to the IL12RB1 gene, which encodes the beta-1 chain of the IL-12/IL-23 receptor, with no expression of the protein in the affected homozygous children (Figure 1A) (20). The homozygous mutation in IL12RB1 (1791+2 T>G) was determined at birth in P1 by genomic DNA sequencing of a cord blood sample. P1 received vaccines for hepatitis B, DPT, measles, mumps, and rubella without any adverse effects, but she did not receive the BCG vaccine and grew healthy without suffering infections. When P1 was 8 years old, she had a mild oral C. albicans infection, which was treated and controlled with topical nystatin. Following a second episode of oral candidiasis, she was treated with the IgG mouthwash procedure as described below.

Patient 2 (P2) was an 8-year-old girl diagnosed with a heterozygous de novo T385M missense mutation in the STAT1 gene that was previously reported as a GOF mutation (Figure 1B) (21). P2 had recurrent and persistent oral Candida infections beginning at 8 months of age. She had received multiple drug treatments with fluconazole, nystatin, ketoconazole, and miconazole, with only partial improvements lasting from one to 3 weeks. At 8 years of age, after receiving unsuccessful treatments for oral candidiasis with nystatin and ketoconazole, she presented with severe oral candidiasis and was treated for 12 days with polyvalent IgG mouthwash.

P1 and P2 were treated on an ambulatory basis with human IV IgG (CSL Behring, PA, U.S.A) according to the following regimen. A 2-min mouthwash was performed with IV IgG solution containing 50 milligrams of IgG diluted in 10 mL of sterile water three times a day after meals and after thorough brushing of the teeth over a 12-day period. After overnight fasting, a daily specimen of early first mouthwash with 5 mL of sterile water was obtained in the morning prior to the brushing of the teeth. The sample was subsequently plated on Sabouraud agar media, and the number of Candida colonies was quantified as colony-forming units (CFU) at 24 h. On day 13, P1 continued to receive the IV IgG (as a mouthwash under the same scheme) and simultaneously started to receive 100,000 IU of nystatin suspension three times a day; IV IgG was administered topically until day 19, while nystatin was administered until day 23 (10 days of treatment with nystatin). P2 stopped the polyvalent human IgG schedule on day 13 when she was admitted to the hospital for treatment for a lower respiratory infection for which she received i.v. amoxicillin and caspofungin for 10 days.

Because the production of oxygen metabolites is a main mechanism by which phagocytic cells kill intracellular microorganisms, we tested the capacity of Hk C. albicans with and without opsonization to induce NADPH oxidase activation in vitro. The rhodamine MFI, which was proportional to the NADPH oxidase activity, was augmented following the in vitro activation of leukocytes. In all experiments, the NADPH activity induced by PMA produced the highest rhodamine MFI as shown in Figure 2 in the upper panel, demonstrating the good cell viability and adequate performance of the test. Although both monocytes and neutrophils gated on FSC vs. SSC graphics showed NADPH oxidase activity in response to the Candida albicans stimulation (data not shown), we focused our analysis on the neutrophils. In all healthy donors tested, the NADPH activity was more enhanced in response to the opsonized C. albicans compared to that in response to the unopsonized preparations as shown in the upper panel of Figure 2. A comparative analysis of the NADPH oxidase activity under the activation conditions was performed using the MFI indices as described in the materials and methods. C. albicans opsonization using 1.5 or 15 mg/mL IgG significantly increased the NADPH activity (median of 5.2- and 6.8-fold with respect to unopsonized Candida) in all healthy donors tested. Similarly, the opsonization with HIS increased the NADPH activity by 3.7- and 4.4-fold when serum was used at 3 or 30%, respectively (Figure 2, lower panel) (p < 0.001). These results show that compared with unopsonized C. albicans, the antibody opsonization of C. albicans with polyvalent commercial human IgG or human serum clearly increases the NADPH oxidase activity of blood neutrophils in healthy human volunteers.

In addition to assessing the NADPH oxidase activity produced by antibody-opsonized C. albicans and comparing with unopsonized C. albicans, we investigated the actual killing activity of the fungus by blood leukocytes from healthy donors to determine whether opsonization helped to eliminate Candida.

The assessment of C. albicans killing using a whole blood assay has the advantage of preserving human phagocytes in their media under a more physiological condition without introducing the stress of cell purification. The experimental conditions of the assays (after the preliminary experiments) were established with a 1-h infection of whole blood with a fixed amount of C. albicans yeast in a shaker, followed by hypotonic cell lysis and seeding of biological material; the CFU were assessed at 24 h.

The killing activity performed by cells in whole blood (putatively by neutrophils and monocytes) against unopsonized Candida albicans was variable among the samples from the 9 healthy donors tested, resulting in a median CFU of 18.2 × 10³ in the experiments (range, 11–26 × 10³ CFU). When Candida was opsonized with IV IgG and tested in the experiments, the number of CFUs at the endpoint decreased in all donors (median, 14.6 × 10³; range, 10.5–19 × 10³), and when Candida was opsonized with IHS, the number of CFUs decreased further (median, 12.8 × 10³; range, 7–18.5 × 10³) (p = 0.05 and 0.003, respectively, Figure 3). We also tested C. albicans killing using a pool of commercial sera obtained from blood type AB positive donors for opsonization and found an adjuvant effect that produced an enhanced killing capacity between that reached with IV IgG and that with immune human sera (data not shown). These significant decreases in the CFU compared with unopsonized Candida and Candida opsonized with IV IgG or immune sera indicate that there was an increase in the killing activity of blood leukocytes in these experiments, which was likely enhanced by the antibodies.

The results of the in vitro experiments with the healthy volunteer blood samples showed that C. albicans opsonization with IV IgG augmented the NADPH and killing activity of phagocytes. On the basis of these results, we tested in vivo the effects of the opsonization of C. albicans with IV IgG in two patients with oral candidiasis and primary immunodeficiency (PID) and observed favorable results. The first patient, P1, who had an IL-12Rβ1 deficiency and C. albicans oral infection, exhibited a positive response to the IgG mouthwash as the number of CFUs of C. albicans in the first daily mouthwash (collected in the morning before eating and brushing teeth) dramatically decreased by 10-fold after only 2 days of treatment (Figure 4A, upper panel). The C. albicans CFU counts continued to decrease over the following days, and on days 11, 15, 16, and 19, the CFUs were null (IgG treatment lasted 19 days for P1). The mouth images of P1 (Figure 4A, lower panel) showed a clear improvement as follows: at the beginning of treatment (day 0), there was intense erythematous candidiasis in the tongue and pseudomembranous candidiasis with white plaques on the tongue and oropharynx; by day 5, the pseudomembranous candidiasis disappeared, and by days 8 and 9, the pseudomembranous candidiasis on the tongue and oropharynx disappeared, which is consistent with the results of the cultures as shown in the plot. Complementary treatment with oral nystatin suspension for 10 days starting on day 13 completely removed the clinical signs of infection, and the patient stopped all treatments. P1 has been free of oral candidiasis for more than 1 year and a half as of the submission of this paper.

P2 has a T385M heterozygous STAT1 GOF mutation (21), and her cells did not produce IL-17 upon the PMA+ionomycin stimulation (Figure 1, lower panels B and C); furthermore, she had CMC refractory to treatment. P2 first presented with severe oral candidiasis affecting the tongue, palate, and lips. On day 0, P2 presented with thick, white crude plaques on the dorsal surface of the tongue, lips, palate, and oropharynges (Figure 4B, first picture). P2 started receiving treatment with only the IV IgG mouthwash under the same schedule as P1. The clearance of Candida infection in her mouth improved as shown in the daily images. For logistical reasons, it was not possible to follow the control of infection by microbiologic cultures as was performed for P1. P2 showed a clear improvement in her mouth following the IV IgG mouthwash treatment, and after 1 week, the appearance of tongue, palate, and lips was substantially clearer after receiving the IV IgG (Figure 4B, day 6) with the resolution of halitosis. The improvement in the oral lesions in P2 allowed her to begin eating better after 1 week; before treatment, eating was painful, and very difficult to perform. On day 14, when the patient was admitted to the hospital for treatment for a lower airway infection, the mouth Candida infection had improved by ~70% compared with that at the beginning of the treatment as evidenced by images. An oral water wash obtained on day 13 was cultured, and C. albicans was identified resistant to voriconazole. In addition, an oral coinfection with Streptococcus sp. was also found in the culture. During the hospitalization, P2 received intravenous amoxicillin and caspofungin, and after 3 days of this treatment, her mouth was free of Candida as evidenced by negative culture. After 1 year and a half receiving the IV IgG mouthwash treatment, P2 only had mild episodes of oral candidiasis, which responded to nystatin, and subsequently, she ceased having severe oral candidiasis.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.