Indian Journal of Pharmacology Home 

[Download PDF]
Year : 2022  |  Volume : 54  |  Issue : 5  |  Page : 338--344

Evaluation of peripheral lymphocyte subsets in acne vulgaris patients before and after systemic isotretinoin treatment

Sibel Dogan Gunaydin1, Ilhan Tezcan2,  
1 Department of Dermatology and Venereology, Hacettepe University, Ankara, Turkey
2 Department of Child Health and Diseases, Division of Pediatric Immunology, School of Medicine, Hacettepe University, Ankara, Turkey

Correspondence Address:
Sibel Dogan Gunaydin
No. 9 Nolu Kapi 2. Kat Hacettepe, Eriskin Hastanesi, Sihhiye 06100, Ankara


OBJECTIVES: Isotretinoin (ISO) is a retinoic acid-derived molecule which is very efficient in the treatment of acne vulgaris (AV). Little is known regarding alterating affects of systemic ISO on immune system. SUBJECTS AND METHODS: Thirty-two patients with moderate and severe AV with indications for systemic ISO treatment were recruited. The evaluation of peripheral blood lymphocyte subsets was made on treatment initiation and at the end of the therapy by flow cytometry. RESULTS: CD16/56+ natural killer (NK) cell levels were higher in patients with severe acne compared to moderate acne (P = 0.009). CD45RA+ (naive) cell levels decreased significantly after systemic ISO treatment; the mean percentage was 66.9 ± 6.5 which decreased to 63.9 ± 7.5 after treatment (P = 0.013). CD19+ cell levels were also significantly increased; mean percentage (%) and absolute cell counts were 8.0 ± 3.1 and 161.3 ± 7.36 cells/μl which increased to 10 ± 2.7 and 227.7 ± 108.7 cells/μl, at the end of treatment, respectively (P = 0.007, P = 0.011). CONCLUSIONS: NK cell count may be a parameter related to acne severity. Systemic ISO may have an inhibitory effect on naïve T cells. Finally, systemic ISO seems to stimulate mature B-cell proliferation.

How to cite this article:
Gunaydin SD, Tezcan I. Evaluation of peripheral lymphocyte subsets in acne vulgaris patients before and after systemic isotretinoin treatment.Indian J Pharmacol 2022;54:338-344

How to cite this URL:
Gunaydin SD, Tezcan I. Evaluation of peripheral lymphocyte subsets in acne vulgaris patients before and after systemic isotretinoin treatment. Indian J Pharmacol [serial online] 2022 [cited 2023 Mar 25 ];54:338-344
Available from:

Full Text


Acne vulgaris (AV) is an inflammatory disease of the follicular unit which is classified as mild, moderate, or severe (nodulocystic) acne according to the occurrence of characteristic skin lesions consisting of comedones, papules, pustules, nodules, and cysts. Isotretinoin (ISO) is a retinoid-derived molecule considered the most efficacious systemic therapeutic option for moderate-to-severe AV. It has superior efficacy on all of the four major etiologic factors in acne pathogenesis; it reduces follicular keratinization, decreases Propionobacterium acnes (P. acnes) colonization, inhibits sebum secretion, and diminishes inflammation.[1],[2] Systemic retinoids are known to have immunomodulatory and anti-inflammatory properties effecting various cell groups in the peripheral circulation.[2] However, it has not been fully understood by which mechanisms systemic ISO suppresses inflammation. The effects of systemic ISO on peripheral lymphocyte subsets, on the other hand, are not yet known. The objective of this study is to evaluate before and after levels of peripheral lymphocyte subsets in AV patients treated with ISO. It is aimed to obtain clear prospective information regarding effects of immunomodulation by systemic ISO treatment on peripheral lymphocytes.

 Subjects and Methods

This is a prospective clinical observational study conducted with 32 patients with AV who had indications for systemic ISO treatment with a dosage of 0. 5 mg/kg/day and continued the treatment for 6 months. Patients (>18 years) who were admitted to Hacettepe University, Department of Dermatology and Venereology and diagnosed…The approvals for the research were obtained from Hacettepe University Ethics Committee for Clinical Researches (no. 2017/12–40 KA 17138) and Turkish Medicines and Medical Devices Agency (no. 18-AKD-42 for Clinical Researches). The research was supported by Hacettepe University Scientific Research Project Coordination Unit number: THD/2017/11120 diagnosed with moderate-to-severe AV along with planned systemic ISO treatment were recruited. Cases who had any kind of medical condition which would contraindicate ISO treatment (i.e., liver function test abnormalities, fasting serum lipid profile abnormalities, psychiatric diseases and depression, pregnancy or lactation, and history of hypersensitivity reaction to the drug or its derivatives); patients with systemic inflammatory or infectious diseases, any kind of neoplastic, infectious or inflammatory diseases of the gastrointestinal system (Crohn's disease and ulcerative colitis); patients who used systemic drugs, cases who received any kind of systemic acne treatment during the past 3 months or had previously been treated with retinoic acid derivatives (ISO, acitretin, and bexarotene); and patients who smoke or use alcohol were excluded.

Demographic data (age, gender, etc.), medical history, and dermatological findings of the patients were noted. Severity of the disease was assessed using the Global Acne Grading System scores (GAGS). In the assessment process, acne lesions in 6 locations including face (chin + nose + cheeks+, forehead), chest, and back were examined. GAGS was calculated by the addition of six local scores derived from multiplying the assigned variable for the most severe lesion within each region (1: ≥one comedone, 2: ≥one papule, 3: ≥one pustule, and 4: ≥one nodule) with area factors (one: Nose, one: Chin, two: Forehead, two: Each check, three: Chest, and three: Back) varying between 0 and 44. The severity of acne in patients was classified into five categories: None (0), mild (1–18), moderate (19–30), severe (31–38 points), and very severe (>39) acne.[2]

Blood samples of the patients were taken twice; first within 15 days after treatment initiation and second in the 6th month of treatment which were subjected to flow cytometric analysis for the investigation of peripheral lymphocyte subsets.

Flow cytometric studies

A volume of 2–4 ml of venous blood sample from each patient was collected in ethylenediaminetetraacetic acid tubes twice which were delivered to the laboratory of the Immunology Unit in Hacettepe University Department of Child Health and Diseases for flow cytometric analysis.

Monoclonal antibodies

CD3/CD16/CD56 monoclonal antibody (M2AB, C5.9, J5511), fluorescein isothiocyanate, PE (catalog number: MA1–12207), MS anti-HU CD45 PERCP (catalog number: MHCD4531), anti-human CD19 APC SJ25C1 (catalog number: 17-0198-41), human CD3/CD4/CD8 mixture CD3/CD4/CD8 (catalog number: CD348A), anti-human CD38 APC-eFluor® 780 (catalog number: 47-0389-41), anti-human HLA-DR APC (LN3) (catalog number: 17-9956-41), anti-human CD45RA APC HI100 (catalog number: 17-0458-41), and anti-human CD45RO APC-eFluor® 780 (catalog number: 47-0457-41) were obtained from Thermo Fisher Scientific (Meridian Road Rockford, IL 61105 USA).

Flow cytometric preparation and assessment

During preparation for flow cytometry, 100 μl of blood was stained with 20 μl of conjugated monoclonal reagent by pipetting. Cells were incubated for in dark for 15 min at 18°C –23°C. 15% formaldehyde and 50% diethylene glycol buffered 2 ml of flow cytometry lysing solution (FACS) lysing solution was added and immediately vortexed. Following incubation for 10 min at 18°C –23°C, the cells were centrifuged at 300 × g for 5 min at room temperature. After aspirating supernatant, 50 μl of the remaining liquid was separated. Resuspension of cells was made in 1% paraformaldehyde and 500 μl phosphate-buffered saline. Samples prepared were kept in the dark at 2°C –8°C until the analysis was commenced. Cells in all fixed samples were analyzed within 24 h. “Attune® Acoustic Focusing Cytometer Applied Biosystems” flow cytometry and the operating system for the device, Attune Cytometer Software 2.1 Thermo Fisher Scientific, Waltham, Massachusetts, USA, were used for evaluations. A panel was designed in the device with preferred antibodies and fluorochromes.

Statistical analysis and ethics

Statistical analysis was performed with Statistical Package for the Social Sciences (SPSS Inc., Chicago, Illinois, USA) for Windows V21.0. “Student's t-test” and “Mann–Whitney U- test” were used for comparisons. The analysis of categorical variables was conducted with “Chi-squared” test and “Fisher's Exact Test.” Ninety-five percent confidence interval was used in analysis and statistically significant results were accepted when P < 0.05. Multiple linear regression was used to model the data generated.

The approvals for the research were obtained from Hacettepe University Ethics Committee for Clinical Researches (no. 2017/12–40 KA 17138) and Turkish Medicines and Medical Devices Agency (no. 18-AKD-42 for Clinical Researches). The research was supported by Hacettepe University Scientific Research Project Coordination Unit number: THD/2017/11120.


Forty patients were included in the study; six patients stopped the treatment due to mucocutaneous side effects, 1 discontinued the treatment achieving an early complete remission and one patient continued his treatment at another facility. The study was completed with 32 patients, 21, 9% (n = 7) was male and 78, 1% (n = 25) was female. The mean age of the patients was 21.7 ± 3. 3 years; age at disease onset was between 15 and 20 years (mean: 18.2 ± 3.7; median: 16) and the duration of complaints ranged between 4 and 12 years (mean: 6.4 ± 1.2; median: 5).

The mean cumulative dose of systemic ISO was 6819.7 ± 1520.7 mg (median: 6200 mg, minimum: 4800 mg, and maximum: 11600 mg). According to disease severity, 40. 6% of the patients (n = 13) had moderate and 59, 4% of the patients (n = 19) had severe acne. Male patients showed a significantly higher rate of severe acne than females (P = 0.013). Cumulative ISO dosage was significantly higher in severe acne patients when compared to those with moderate acne (P = 0.034) [Table 1].{Table 1}

The mean GAGS of the patients was 30.3 ± 4.9 which decreased to 6.2 ± 3.5 after treatment. The difference between mean GAGS before and after treatment was statistically significant in all patient population, in severe AV and also in patients moderate AV, respectively (P < 0.01 for all three comparisons).

Percentage (%) and absolute cell count of lymphocyte subsets CD45, CD19, CD45RA, CD45RO, CD3, CD4, CD8, CD38, HLADR, CD16/56+ were investigated in all patients and were found within reference ranges. Patients in each acne severity group were also compared with respect to cell percentages and absolute cell counts of the lymphocyte subsets, complete blood count, liver function test, and fasting serum lipid profile tests. Absolute CD16/56+ cell counts of severe acne group were higher; mean CD16/56+ cell percentage and absolute cell count were 12.6 ± 6.1 and 299.6 ± 200.2 cells/μl, respectively. These values were 9.5 ± 3.0 and 183.6 ± 83.6 cells/μl in moderate acne group which were found as significantly lower statistically, respectively (P = 0.098, P = 0.048). There was no significant difference between two AV groups in terms of other flow cytometry variables [Table 2].{Table 2}

Pre-treatment GAGS were found to be positively correlated with cell percentages of CD45RA+ and CD16/56+ and negatively correlated with cell percentages of CD3+ and CD4+ (P = 0.009; r = 0.453; P = 0.022, r = 0.440; P = 0.001, r = −0.541; P = 0.014, r = −0.430). There was no statistically significant relation between GAGS and cell percentages of other lymphocyte subsets (P > 0.05). Pretreatment GAGSs were positively correlated with absolute cell count of CD45RA+ and CD16/56+ as well (P = 0.018, r = 0.417; P = 0.009, r = 0.491). There was no statistically significant relation between GAGS and absolute cell counts of other lymphocyte subsets (P > 0.05). A regression analysis model showed these variables were observed to have an impact on acne severity at the rate of 79.5% for CD16/56, 47.7% for CD45RA, 61.5% for CD3, and 57.3% for CD4 together with male gender. On the other hand, none of the significantly correlated variables had an effect on the acne severity per se with a correction for gender.

In comparisons between before and after treatment evaluations in all patients; mean CD45RA+ cell percentage which was 66.9 ± 6.5 before treatment decreased down to 63.9 ± 7.5 and the difference was found statistically significant (P = 0.013). Basal mean CD19+ cell % and absolute cell count were 8.0 ± 3.1 and 161.3 ± 73.6 cell/μl, these values significantly increased to 10 ± 2.7 and 227.7 ± 108.7 cell/μl after treatment.(P = 0.007, P = 0.011) [Table 3] and [Figure 1] and [Figure 2]. The mean CD19+ cell percentage increased from 8.0 ± 2.6 to 10.6 ± 3.4 and mean CD45RA+ cell percentage decreased from 62.2 ± 6.2 to 60.0 ± 6.5 after the treatment in moderate AV group. The differences were statistically significant for both parameters, respectively (P = 0.033, P = 0.04). There was not a statistically significant difference in comparisons of peripheral lymphocyte subsets in severe AV group (P > 0.05 for all comparisons).{Table 3}{Figure 1}{Figure 2}

There were statistically significant positive correlations between cumulative dose of ISO and GAGS before treatment, cumulative dose of ISO and GAGS scores after treatment, and GAGS scores before and after treatment (P < 0.001, r = 0.668; P = 0.042, r = 0.374; P = 0.001, r = 0.582). A cumulative dose of ISO administered during treatment was positively correlated with posttreatment CD16/56+ cell percentages and negatively correlated with CD45RO+ cell percentages and absolute cell counts (respectively; P = 0.04, r = 0.599; P = 0.032, r = −0.618; P = 0.0017, r = −0.672). A regression analysis revealed these variables were observed to have an effect on the cumulative dose of ISO at the rate of 84,6% for CD45RO and 79,6% for CD16/56 together with male gender. However, none of the significantly correlated variables showed an effect on the cumulative dose of ISO per se when correction for gender was made.


AV is an inflammatory skin disease where immune responses of the host may trigger and perpetuate severe cutaneous reaction.[1],[2] Thus, baseline peripheral lymphocyte subsets and their relation to acne severity were a point of interest in our study, where CD16/56+ cell counts were higher in severe acne patients. Pretreatment GAGS of patients were also positively correlated with CD16/56+ cell percentages and absolute cell counts. Moreover, slight decrease in CD16/56+ cell levels was observed after treatment. The established range of references for peripheral T-cell subsets in healthy population shows no significant difference males and females in terms of T, B, and natural killer (NK) cell markers levels.[3] On the other hand, females are also reported to have higher CD4+ T and memory T lymphocytes and lower CD16+ cell counts.[4] Similar inter-gender differences are seen in other studies conducted with a variety of populations.[5],[6] In our study, male patients outnumbered females in severe acne group which was expected since acne is generally accepted to be more severe in male gender, regression models were used to assess the effects of the gender on the comparative statistical analyses of the patients according to their disease severity. This analysis also revealed that CD16/56+ cell values, together with male gender, were related with acne severity at a rate of 79.5%. Holland et al. conducted the first study investigating peripheral lymphocyte subsets in AV where patients with various disease severity were compared with controls which revealed T helper lymphocytes and B lymphocytes increased with acne severity.[7] In another study, 15 patients with severe nodular acne or acne fulminans were assessed in different acne phases with luminol-enhanced chemiluminescence method where no significant alterations were observed in the percentages of T helper cells, HLA-DR+ lymphocytes, and NK cells or in soluble Interleukin-2 receptor levels.[8] In another study carried out with patients with synovitis, acne, pustulosis, hyperostosis, osteitis syndrome, 19 patients with healthy controls were compared; NK cells were lower, and Th17/Treg cells were higher in the patient group.[9] Although the results in the literature are incompatible with each other, this is attributed to the limitation of these studies to small patient groups with heterogeneous disease activity and use of variable methodologies for measurements. We assume that to explore a possible relation between acne severity and CD16/56+ NK cells, studies including comparison of larger number of subjects including homogeneous gender distribution with strictly defined acne severity groups are required.

The primary pathogenic mechanism of acne is inflammatory response to the pathogens in normal skin flora. P. acnes stimulates the innate immune system through Toll-Like Receptors-2 (TLR-2) expression, excessive inflammatory response is observed as the severity of acne increases.[10],[11],[12] Similarly, mycobacterial cell wall components interact with TLR2 in an in vitro setting, lead to NK cell activation and IFN-γ stimulation.[13] TLR-2 polymorphisms are suggested to change pulmonary tuberculosis phenotypes by influencing the expansion of NK cells.[14] Our findings regarding NK cells and acne severity may also hint a possible association of TLR-mediated immune response to P. acnes resulting in NK cell proliferation in AV. The existence of such a pathomechanism is noteworthy and should be documented by studies that will investigate NK cells together with TLR polymorphism in AV patients.

Inflamed acne lesions are shown to contain high numbers of CD4+, HLA-DR+ activated T cells, where CD45RA+ (naive) and CD45RO+ (memory) T cells consist 50% of the cell population.[15] Although naive T cells are found in increased numbers in inflamed lesional skin, it is not yet known whether this affects peripheral blood levels in proportion with associating acne severity. The existence of this possible relation, on the other hand, is supported by our findings revealing a positive correlation present between CD45RA+ cells and GAGS and a significant decreased of CD45RA+ cells after systemic ISO.

One of the most interesting results of our study was the increase in CD19+ cells in AV patients after systemic ISO. CD19 molecule, which is also known as “B Lymphocyte Surface Antigen B4,” is a cell surface molecule rooted in B lineage cells and is expressed in all B cells except for plasma cells.[16] It is known that there is significantly increased CD19 expression in mature B cells when compared to immature B cells.[17] In in vitro experiments, retinoid derivatives show different affects on immunoglobulin production and B-cell proliferation which is attributed to their varying structural functions.[18] There are also studies indicating that retinoic acid can normalize B-cell functions in common variable immunodeficiency (CVID) patients.[19],[20] In a study conducted with CVID, patients were administered 0.5 mg/kg 13-cis retinoic acid for 12 weeks, B cells displayed a more differentiated cell surface phenotype revealing that retinoic acid may induce functional differentiation.[20] Retinoic acid derivatives are also effective in the inhibition of B lymphocytes that are immortalized by Epstein–Barr virus by their multiple effects on G1 regulator proteins in the cell cycle and by reducing pRb phosphorylation.[21] Foreign CpG-DNA structures from pathogens activate memory B cells through TLRs, and retinoic acid is shown to increase CpG-meditated stimulation of the proliferation of peripheral B cells.[22] Eriksen et al. investigated effects of all-trans retinoic acid (ATRA) on B cells through TLR homolog RP105 molecule, where ATRA was shown to be a strong stimulator with prior CpG oligomer application to cell cultures. ATRA had specifically strong impacts on memory B cells immunoglobulin secretion consequently revealing TLR signaling had an important role in inducing ATRA as a potent RP105-mediated B-cell stimulator.[23] In a study investigating ATRA's TLR-mediated effects in patients with CVID, it was found that RA strengthened proliferation of B cells and expression of CD80 activation marker.[24] Considering the demonstrated effects of ATRA on B cells in the presence of TLR activation, our results regarding CD19+ cells were interpreted significant and assumed that a reinforced effect due to the use of systemic ISO for acne causes this increase. It can, therefore, be argued that increased interaction between P. acnes and TLR2 in moderate and severe AV and additional systemic ISO treatment may lead to the increase in CD19+ cell percentage.


Despite a small number of patients, clinical evidence for peripheral immunomodulation with systemic ISO at AV treatment doses was able to be shown in this study. It will be possible to obtain further information on this subject by supporting our results with studies carried out with higher number of patients.

Financial support and sponsorship

The research was supported by Hacettepe University Scientific Research Project Coordination Unit number: THD/2017/11120.

Conflicts of interest

There are no conflicts of interest.


1Bhat YJ, Latief I, Hassan I. Update on etiopathogenesis and treatment of Acne. Indian J Dermatol Venereol Leprol 2017;83:298-306.
2Zaenglein AL, Pathy AL, Schlosser BJ, Alikhan A, Baldwin HE, Berson DS, et al. Guidelines of care for the management of acne vulgaris. J Am Acad Dermatol 2016;74:945- 73.e33.
3Valiathan R, Deeb K, Diamante M, Ashman M, Sachdeva N, Asthana D. Reference ranges of lymphocyte subsets in healthy adults and adolescents with special mention of T cell maturation subsets in adults of South Florida. Immunobiology 2014;219:487-96.
4Rudy BJ, Wilson CM, Durako S, Moscicki AB, Muenz L, Douglas SD. Peripheral blood lymphocyte subsets in adolescents: A longitudinal analysis from the REACH project. Clin Diagn Lab Immunol 2002;9:959-65.
5Shahal-Zimra Y, Rotem Z, Chezar J, Shochat T, Ross L, Pickholtz I, et al. Lymphocyte subset reference ranges in healthy Israeli adults. Isr Med Assoc J 2016;18:739-43.
6Al-Thani A, Hamdi WS, Al-Marwani A, Alnaqdy A, Sharafeldin H. Reference ranges of lymphocyte subsets in healthy Qatari adults. Biomark Med 2015;9:443-52.
7Holland DB, Gowland G, Cunliffe WJ. Lymphocyte subpopulations in patients with AV. Br J Dermatol 1983;109:199-203.
8Karvonen SL, Räsänen L, Soppi E, Hyöty H, Lehtinen M, Reunala T. Increased chemiluminescence of whole blood and normal T-lymphocyte subsets in severe nodular acne and acne fulminans. Acta Derm Venereol 1995;75:1-5.
9Xu D, Liu X, Lu C, Luo J, Wang C, Gao C, et al. Reduction of peripheral natural killer cells in patients with SAPHO syndrome. Clin Exp Rheumatol 2019;37:12-8.
10Leyden J. How does our increased understanding of the role of inflammation and innate immunity in acne impact treatment approaches? J Dermatolog Treat 2016;27 Suppl 1:S1-3.
11Valins W, Amini S, Berman B. The expression of toll-like receptors in dermatological diseases and the therapeutic effect of current and newer topical toll-like receptor modulators. J Clin Aesthet Dermatol 2010;3:20-9.
12Kistowska M, Gehrke S, Jankovic D, Kerl K, Fettelschoss A, Feldmeyer L, et al. IL-1b drives inflammatory responses to Propionibacterium acnes in vitro and in vivo. J Invest Dermatol 2014;134:677-85.
13Esin S, Counoupas C, Aulicino A, Brancatisano FL, Maisetta G, Bottai D, et al. Interaction of Mycobacterium tuberculosis cell wall components with the human natural killer cell receptors NKp44 and Toll-like receptor 2. Scand J Immunol 2013;77:460-9.
14Chen YC, Hsiao CC, Chen CJ, Chin CH, Liu SF, Wu CC, et al. Toll-like receptor 2 gene polymorphisms, pulmonary tuberculosis, and natural killer cell counts. BMC Med Genet 2010;11:17.
15Holland DB, Jeremy AH, Roberts SG, Seukeran DC, Layton AM, Cunliffe WJ. Inflammation in acne scarring: A comparison of the responses in lesions from patients prone and not prone to scar. Br J Dermatol 2004;150:72-81.
16Tedder TF, Isaacs CM. Isolation of cDNAs encoding the CD19 antigen of human and mouse B lymphocytes. A new member of the immunoglobulin superfamily. J Immunol 1989;143:712-7.
17Scheuermann RH, Racila E. CD19 antigen in leukemia and lymphoma diagnosis and immunotherapy. Leuk Lymphoma 1995;18:385-97.
18Dillehay DL, Jiang XL, Lamon EW. Differential effects of retinoids on pokeweed mitogen induced B-cell proliferation vs. Immunoglobulin synthesis. Int J Immunopharmacol 1991;13:1043-8.
19Saxon A, Keld B, Braun J, Dotson A, Sidell N. Long-term administration of 13-cis retinoic acid in common variable immunodeficiency: Circulating interleukin-6 levels, B-cell surface molecule display, and in vitro and in vivo B-cell antibody production. Immunology 1993;80:477-87.
20Adelman DC, Yen TY, Cumberland WG, Sidell N, Saxon A. 13-cis retinoic acid enhances in vivo B-lymphocyte differentiation in patients with common variable immunodeficiency. J Allergy Clin Immunol 1991;88:705-12.
21Zancai P, Cariati R, Rizzo S, Boiocchi M, Dolcetti R. Retinoic acid-mediated growth arrest of EBV-immortalized B lymphocytes is associated with multiple changes in G1 regulatory proteins: P27KIP1 up-regulation is a relevant early event. Oncogene 1998;17:1827-36.
22Ertesvag A, Aasheim HC, Naderi S, Blomhoff HK. Vitamin A potentiates CpG-mediated memory B-cell proliferation and differentiation: Involvement of early activation of p38MAPK. Blood 2007;109:3865-72.
23Eriksen AB, Indrevær RL, Holm KL, Landskron J, Blomhoff HK. TLR9-signaling is required for turning retinoic acid into a potent stimulator of RP105 (CD180)-mediated proliferation and IgG synthesis in human memory B cells. Cell Immunol 2012;279:87-95.
24Indrevær RL, Holm KL, Aukrust P, Osnes LT, Naderi EH, Fevang B, et al. Retinoic acid improves defective TLR9/RP105-induced immune responses in common variable immunodeficiency-derived B cells. J Immunol 2013;191:3624-33.