Clickable Vitamins as a New Tool to Track Vitamin A and Retinoic Acid in Immune Cells

Abstract

The vitamin A derivative, retinoid acid (RA) is key player in guiding adaptive mucosal immune responses. However, data on the uptake and metabolism of vitamin A within human immune cells has remained largely elusive because retinoids are small, lipophilic molecules which are difficult to detect. To overcome this problem and to be able to study the effect of vitamin A metabolism in human immune cell subsets, we have synthesized novel bio-orthogonal retinoid-based probes (clickable probes), which are structurally and functionally indistinguishable from vitamin A. The probes contain a functional group (an alkyne) to conjugate to a fluorogenic dye to monitor retinoid molecules in real-time in immune cells. We demonstrate, by using flow cytometry and microscopy, that multiple immune cells have the capacity to internalize retinoids to varying degrees, including human monocyte-derived dendritic cells (DCs) and naïve B lymphocytes. We observed that naïve B cells lack the enzymatic machinery to produce RA, but use exogenous retinoic acid to enhance CD38 expression. Furthermore, we showed that human DCs metabolize retinal into retinoic acid, which in co-culture with naïve B cells led to of the induction of CD38 expression. These data demonstrate that in humans, DCs can serve as an exogenous source of RA for naïve B cells. Taken together, through the use of clickable vitamins our data provide valuable insight in the mechanism of vitamin A metabolism and its importance for human adaptive immunity.

Keywords: CD38; copper-facilitated click chemistry; retinoic acid; retinoid probes; vitamin A metabolism.

Scheme 1

Synthesis of clickable vitamins. Reagents and conditions: a) n-Buli, THF, -78°C to RT, 3 h, 8% for RE-click and 39% for 4; b) 2M NaOH, MeOH, 50°C, 4 hours, 24%.

Scheme 2

Synthesis of mono-protected aldehyde intermediate. Reagents and conditions: a) tBuOCl, AcOH, 1h and then CuSO₄, H₂SO₄, 96h, 30%; b) Na₂CO₃, MeOH/H₂O, 5h, 0°C, 33%; c) PCC, DCM, 90 min, 0°C, 51%; d) trimethyl orthoformate, pTsOH, MeOH, 24h, 67%; e) K₂HPO₄/KH₂PO₄, NaBr, DMSO, 24h, 80°C, 55%.

Figure 1

Monocyte-derived dendritic cells have the capacity to take up and metabolize retinoids. (A) Chemical structure of retinal and its synthetic clickable counterpart which contains an alkyne ligation group to allow visualization. (B) Chemical structure of retinoic acid and its synthetic clickable couterpart which contains an alkyne ligation group to allow visualization. (C–K) Monocytes were isolated from blood and differentiated into monocyte-derived dendritic cells (moDCs) during 7 days of culture in the presence of GM-CSF and IL-4. (C) Representative dot plots of aldeluor positivity in moDCs and DEAB-treated negative control. (D) Quantification of the aldefluor positivity in moDCs from 6 different donors. (E) Quantification of the increase of aldefluor positivity in moDCs compared to the paired DEAB control from 7 different donors. (F–K) Clickable retinoid probes were incubated for one hour in RPMI at 37°C with differentiated moDCs. (F) Representative histogram of the uptake of RE-click (pink) compared to the vehicle control (dotted line) in moDCs. (G) Quantification of the uptake RE-click compared to the vehicle control in moDCs from 3 different donors. (H) Representative histogram of the uptake of RA-click (blue) compared to the vehicle control (dotted line) in moDCs. (I) Quantification of the uptake of RA-click compared to the vehicle control in moDCs from 4 different donors. (J) Differentiated moDCs were incubated with multiple concentrations of RE­ and RA-click ranging from 1000 to 1 nM. Quantification of the uptake of the clickable probes was detemtined using flow cytometry using blood form 3 different donors (K) moDCs were put on a coverslip to assess the spatial distribution of retinoid probes (green) and nuclei (DAPI, blue) using confocal microscopy. White arrow indicates cytoplasmic accumulation, whereas the red arrow indicates nuclear localization. Data are presented as mean± SD. Student's t test; *p < 0.05.

Figure 2

Uptake of dickable retinoids in human dendritic cells is independent of protein and temperature. Monocytes were isolated from blood and differentiated into monocyte-derived dendritic cells (moDCs) during 7 days of culture in the presence of GM-CSF and IL-4 from 3 different donors. Uptake of (A) RE-click and (B) RA-dick compared to the vehicle control in differentiated moDCs after incubation in phosphate buffered saline (PBS) for one hour at 37°c was determined by flowcytometric analysis and expressed as mean fluorescent intensity (MFI). Uptake of (C) RE-click and (D) RA-click compared to the vehicle control in differentiated moDCs after incubated in phosphate buffered saline (PBS) for one hour at 4°C was determined by flowcytometric analysis and expressed as MFI. (E) Representative histogram of uptake of RE- (pink) and RA-click (blue) compared to the vehicle control (solid line) in moDCs at 4°C, as shown in (C, D) Data are presented as mean± SD. Student's t test; *p < 0.05, **p < 0.01.

Figure 3

Clickable RA probe induces tolerogenic phenotype in monocyte-derived dendritic cells. Monocytes were isolated from blood and differentiated into monocyte-derived dendritic cells (moDCs) during 7 days of culture in the presence of GM-CSF and IL-4 from 3 different donors. (A) Aldefluor positivity and (B) CD103 expression in moDCs differentiated in the presence of natural RA or RA­ click. Data are presented as mean± SD. Student's t test; ANOYA; **p < 0.01, ***p < 0.001.

Figure 4

Human naive B cells cannot produce retinoic acid, but are affected by exogenous retinoids to upregulate CD38 expression. Human naive B cells were isolated from peripheral blood. (A) Representative dot plot of aldefluor positivity in naive B cells. DEAB reagent was used as a negative control. (B) Quantification of aldefluor positivity in naive B cells from 3 different donors. (C) Representative histogram of the uptake of RE-click (pink) and RA-dick (blue) in naive B cells compared to vehicle control (dotted line). (D) Quantification of the uptake of RE-click and RE-click in naive B cells compared to vehicle control from 3 different donors. (E) Visualization of the RE- and RA-click using Azide-AF488 (green) with a nucleus staining (DAPI, blue) in naive B lymphocytes from 5 different donors. (F) Human naive B cells were isolated from peripheral blood and cultured in the presence of T cell dependent (TD) stimuli for 7 days, including interleukin-4, anti-CD40 and anti-IgM antibodies. Histogram of CD38 expression on B cells upon addition of unmodified retinal (green) or retinoic acid {grey) in comparison to vehicle control (dotted line). (G) Histogram of CD38 expression upon culturing with RA-click (blue), its unmodified counterpart retinoic acid (grey) and vehicle control (dotted line) during 7 day TC-dependent stimulation. (H) Quantification of CD38 expression upon culturing with the RA-click, its unmodified counterpart retinoic acid and vehicle control from 4 different donors. Data are presented as mean ± SD. ANOVA; *p < 0.05. ns indicates p > 0.05.

Figure 5

Monocyte-derived dendritic cells convert retinal into retinoic acid, and increase B cell CD38 expression. Naive B cells and monocyte-derived dendritic cells were co-cultured in vitamin A deficient medium in the presence of TD-dependent stimuli for 9 days, including interleukin-4, anti-CD40 and anti-IgM antibodies. As a negative control naive B cells and monocyte-derived dendritic cells were co-cultured with vehicle alone. (A) Representative histogram of CD38 expression on naive B cells supplemented with natural retinal (green) or RA (black) during the 9 day coculture. Vehicle treatment is indicated with the dotted line. As a negative control, co­cultures were treated with vehicle alone (dotted line). (B) Quantification CD38 mean fluorescent intensity of naive B cells supplemented with unmodified retinal (green), RA (black) or vehicle treated cells (white) from 4 separate experiments with individual blood donors (n=4). Data are presented as mean ± SD. ANOVA; **p < 0.01.