Fasting mitigates immediate hypersensitivity: a pivotal role of endogenous D-beta-hydroxybutyrate

All animal experiments were performed under the following conditions; Room temperature 23 ± 2°C, relative humidity 60% ± 10%, and an alternating 12-hours light–dark cycle (8 AM to 8 PM). Animals were quarantined and acclimatized before experiments at least for a 1-week period standard diet (Oriental Yeast Industries, Tokyo, Japan) and water were available ad libitum. The care and handling of the animals were in accordance with the National Institute of Health guidelines. All procedures involving animals were carried out in accordance with the Institutional Guidelines.

For fasting treatment, mice and rats had access to only water. For the ketogenic diet, Diet D10070801 (Research Diets, New Brunswick, NJ, USA), consisting of 67% fat, 17% protein, and 0% carbohydrate was used.

Nasal Allergy models were performed by the method described previously with slight modifications [23]. Six-week-old male Brown Norway rats (n = 6–8 in each experiment) were purchased from Japan SLC, Inc. Five μl of a 10% solution of TDI (Sigma-Aldrich, St. Louis, MO, USA) in ethyl acetate (Sigma-Aldrich, St. Louis, MO, USA) was intranasally applied to the bilateral nasal cavities once a day for five consecutive days (sensitization). This sensitization procedure was then repeated after a 1-day interval. Seven days after the latest sensitization, and prior to being evaluated, 5 μl of 10% TDI solution was applied again to the nasal cavaties to provoke nasal allergy-like behaviors. Sensitized rats without provocation by TDI solution served as the control group.

Nasal allergy-like behaviors were evaluated by counting the amount of nasal rubbing and sneezing during a 20 minute period immediately after TDI application. Changes in nasal secretions were evaluated by measuring the weight of cotton thread 60 minutes after TDI application.

The total plasma IgE levels were determined after TDI application. Blood was collected from the abdominal aorta. Plasma samples were obtained by centrifugation and stored at -30°C until use. The total plasma IgE level was determined by using Rat IgE ELISA Kit (Shibayagi Co., Ltd, Gumma, Japan) according to the manufacturers instruction.

After nasal secretion was measured, rats were applied to histological examination of the nasal cavity and mucosa of maxilloturbinate. Rats were euthanized by cervical dislocation under anesthetized with sodium pentobarbital. The head was separated from the neck and the whole anterior face sections were prepared using a multipurpose cryosection preparation kit (Section Lab Co., Ltd, Hiroshima, Japan) according to the method of Kawamoto [24]. In brief, anterior face of tissue was frozen in isopentane (-160°C) cooled with dry ice and then freeze-embedded with SCEM compound (Section Lab Co., Ltd, Hiroshima, Japan). Twenty-μm thick sections were cut with cryomicrotome (CM3050S, Leica Co., Ltd, Heidelberg, Germany) and collected with cryofilm (Section Lab Co., Ltd, Hiroshima, Japan). For hematoxylin and eosin (H&E) and Toluidine blue (pH4.1) staining, tissue sections were fixed by 4% paraformaldehyde after drying section and mounted with SCMM (Section Lab Co., Ltd, Japan). For immunohistochemistry, the sections were incubated with a polyclonal antibody to mast cell tryptase (Santa Cruz, CA, USA) at 4°C overnight, followed by labeling with Alexa 488-conjugated goat anti-rabbit IgG (Life Technologies, Inc. Grand Island, NY, USA) and Hoechst 33342 (Dojindo Co., Ltd, Kumamoto, Japan) at RT for 30 min. These sections were observed with microscope (BZ-9000, Keyence Co., Ltd., Tokyo, Japan) or a confocal microscope (LSM 710, Carl Zeiss, Inc. Heidelberg, Germany).

Six-week old male Wistar rats (Japan Clea, Inc. Tokyo, Japan) were used (n = 7–27 in each time point). The homologous antiserum used was prepared according to the method described. Briefly, rats were sensitized by intramuscular injection of 0.2 ml of 10 mg/mL egg albumin (Sigma-Aldrich, St. Louis, MO, USA) to both femurs and intraperitoneal administration of Bordetellapertussis inactive microorganism suspension (B. pertussis, Kitasato Institute Research Center for Biologicals, Saitama, Japan). On day 14, animals were bled and antiserum was collected. Separated antiserum was stored at -80°C.

Rat homologous antiserum (100 μL) was injected into the shaved back skin of rats. After 24 h, sensitized rats were injected with 0.5 mL of mixed solution of 0.5% Evan blue (Fluka, Buchs, Switzerland) and 1% egg albumin (1:1) through the tail vein. Thirty minutes later, rats were sacrificed and the dye content of the shaved dorsal skin was determined quantitatively. Pieces of the dorsal skin containing extravasated dye were removed and soaked overnight in stoppard glass tubes containing 1 ml of mixed solution of 0.6 N H₃PO₄/acetone (13:5). The absorbance of supernatant was measured at 620 nm using a spectrophotometer and the concentration of the dye in the shaved dorsal skin (μg/site) was determined. Fasting duration was determined as the time between food removal and antigen challenge.

Ten-week old male BALB/c mice (Japan Clea, Inc. Tokyo, Japan) were used (n = 6–12 in each experiment). For the evaluation for IgE dependent anaphylaxis, mice were sensitized with an intraperitoneal injection of 125 μg egg albumin dissolved in aluminum gel. Seven days later, mice were challenged by intraperitoneal injection of 125 μg egg albumin dissolved in aluminum gel [25]. For IgE independent anaphylaxis, anaphylactic shock was evoked by intraperitoneal injection of 6.5 mg/kg compound 48/80 (Sigma-Aldrich, St. Louis, MO, USA) in saline solution [26]. Immediately after the antigen or compound 48/80 challenge, rectal temperature was measured with a digital thermometer (Tateyama kagaku industry Co.,Ltd, Toyama, Japan).

Eleven-week old male Sprague Dawley rats (Japan Clea, Inc. Tokyo, Japan) were used (n = 4 in each treatment). Rat peritoneal mast cells (RPMC) were collected by rinsing the peritoneal cavities of normal rats with 20 ml of mast cell medium (MCM, containing 150 mM NaC1, 3.7 mM KC1, 3.0 mM Na₂HPO₄, 3.5 mM KH₂PO₄, 5.6 mM dextrose,0.1% gelatin, 0.1% (w/v) bovine serum albumin) containing 10 U/ml heparin under anesthesia with pentobarbital. Peritoneal cells were sedimented twice at 50 x G for 7 minutes at 4°C and re-suspended in MCM. RPMC were then washed twice with siraganian buffer [119 mM NaCl, 5 mM KCl, 0.4 mM MgCl₂, 25 mM piperazine-N,N0-bis(2-ethanesulfonic acid)(PIPES), pH 7.2] and resuspended in siraganian buffer. RPMC (5 × 10⁴).

The degree of degranulation was determined by measuring the release of β-hexosaminidase [27]. The enzymatic activity of β-hexosaminidase in supernatants and sonicated disrupted cell pellets was measured with p-nitrophenyl N-acetyl-β-d-glucosaminide (Sigma-Aldrich, St. Louis, MO, USA) in 0.1 M sodium citrate (pH 4.5) for 60 minutes at 37°C. The reaction was stopped by the addition of 1 M Na₂CO₃ glycine. The release of the product, 4-p-nitrophenol, was detected by absorbance at 405 nm. The extent of degranulation was calculated by dividing 4-p-nitrophenol absorbance in the supernatant by the sum of absorbance in the supernatant and disrupted cell pellet.

To induce RPMC degranulation, cells were incubated with 1 μg/mL compound 48/80 at 37°C for 10 minutes. For the evaluation of the effect of D-BHB on mast cell degranulation, cells were preincubated with various concentrations of D-BHB for 10 minutes, and then incubated with compound 48/80.

The [Ca²⁺]_i in RPMC was measured with the fluorescent Ca²⁺ indicator, Fura-2/AM (Invitrogen, Carlsbad, CA, USA). Cells were plated onto glass bottom culture dishes. The cell was loaded with the indicator by incubation with 2.5 μM of Fura-2/AM in in Standard Solution (140 mM NaCl, 5 mM KCl, 1 mM MgCl₂, 10 mM HEPES, 10 mM dextrose) for 4 h at room temperature (22–24°C). After loading, cells were washed three times with Standard solution and incubated for further 30 minutes to ensure de-esterification. Fluorescence recordings were performed using two set-ups, employing either video-imaging or a single detector system. For video-imaging, a coverslip with dye-loaded cells was mounted in a chamber fixed on the stage of an inverted fluorescence microscope (IX71, Olympus Co., Tokyo, Japan) equipped with UV objective lens (UApo 20X3/340, Olympus Co., Tokyo, Japan). The emission signal passed through a band pass filter (500 ± 10 nm) and was detected by a 12 bit CCD camera (C-8214, Hamamatsu Photonics K.K., Hamamatsu, Japan). Fluorescence intensities at 510 nm with excitation at 340 and 380 nm were recorded at 5 s intervals. The [Ca²⁺]_i in an individual cell was calculated from the ratio of fluorescence images acquired at excitation at 340 nm to those at 380 nm. All experiments were performed at room temperature (22–24°C).

Increases in [Ca²⁺]_i were evoked by 1 μg/mL compound 48/80 for 240 sec. For the evaluation of the effect of D-BHB on [Ca²⁺]_i, cells were stimulated with compound 48/80 under 100 mM D-BHB.

Eleven-week old male Sprague Dawley rats (Japan Clea, Inc. Tokyo, Japan) were used (n = 4 in each treatment). The concentration of D-BHB in collected serum and intraperitoneal fluid was measured using Ketone test A (Sanwa kagaku kenkyusho Co., Ltd., Nagoya, Japan) according to the manufacturers instruction.

We used the Student’s t-test for comparison of the two groups and the Dunnett test for multiple comparisons. The Pearson product–moment correlation coefficient was used to evaluate association between anaphylactic response and changes in serum D-BHB concentration. Differences between measured variables were considered significant if the resultant P value was 0.05 or less. Analysis was performed using the JMP version 8 (SAS Institute Inc., Cary, North Carolina, USA).

Figure 1 shows changes in allergy-like behaviors, nasal rubbing (A), sneezing (B), and nasal secretion (C) and serum IgE concentration (D) induced by 24 hours fasting. For the food and water Ad libitium group (AL), TDI application induced significant increases in each allergy behavior compared to control group. For the 24 hours fasting group, TDI application induced increases in each allergy behavior. Though, the levels were significantly lower than that of AL + TDI group.

Corresponding to previous measurements of sensitization with TDI [28, 29], total serum IgE concentration was significantly higher in AL with TDI group than in the normal value. For the fasting with TDI group, significant increases were observed than the normal value. Though, no significant difference was observed between AL with TDI group and fasting with TDI group.Figure 2 shows mucus secretion and mast cell alteration in the nasal cavity and mucosa of maxilloturbinate. The mucus secretion in the nasal cavity and epithelia was increased in the AL with TDI group compared to control group. In the 24 hours fasting group with TDI, these mucus secretion were maintained as same level as control group (Figure 2 upper).The mast cells presented in lamina propria of nasal epithelia were non-degranulated in the control group. In the AL with TDI group, mast cells were degranulated. Twenty four hours fasting with TDI maintained the mast cells non-degranulated (Figure 2 middle).The content of tryptase in the mast cells was apparent in the control group than AL with TDI group. In the 24 hours fasting group with TDI, mast cell tryptase were maintained as same level as control group (Figure 2 lower). These histological observations support the results of suppressed nasal hypersensitivity reaction by the fasting.

To examine the effect of food restriction on systemic anaphylaxis, we evaluated antigen or compound 48/80 induced anaphylactic responses in immediate type hypersensitivity by reductions in rectal temperature. For the 24 hours fasting group, reductions in rectal temperature were significantly smaller than those of the AL group in both antigen and compound 48/80 induced anaphylaxis models (Figures 3A and B).

It has been known that fasting markedly increases circulating concentrations of D-BHB [11]. To investigate the relationship between anaphylactic response and changes in serum D-BHB concentrations by food restriction, we quantitatively analyzed the anaphylactic response using the PCA reaction. In rats during fasting, PCA was reduced in a time-dependent manner and significant reductions were observed after 12 hours starvation (Figures 4A and B). In contrast to the PCA reaction, serum D-BHB concentrations increased in a time dependent manner and significant increases were observed after 12 hours fasting (Figure 4C). This result almost corresponds to a previous report [30]. A significant correlation between individual PCA and D-BHB concentrations was observed during the fasting period (Figure 4D).

Degranulation of mast cells plays a crucial role in the pathogenesis of anaphylaxis and allergic disorders [22]. Two methods were used to compare the effect of elevated systemic D-BHB and mast cell degranulation with fasting; a ketogenic diet, which accelerates fatty acid oxidation; and systemic instillation of D-BHB.For fasting, significantly (approximately ten-fold) higher serum and intraperitoneal D-BHB concentrations were observed 15 and 24 hours after food restriction than that before food restriction. The elevated D-BHB concentrations returned to pre food deprivation values 3 hours after refeeding with the control diet (Figure 5 upper middle).Corresponding to changes in D-BHB concentrations, mast cell degranulation to compound 48/80 was inhibited 15 and 24 hours after food restriction and returned to pre food deprivation ratios 3 hours after the refeeding period (Figure 5 upper right).For the ketogenic diet, significantly (approximately eight-fold) higher serum and intraperitoneal D-BHB concentrations were observed 15 and 24 hours after food feeding of the ketogenic diet than that before feeding of the ketogenic diet (Figure 5 middle left and middle center). These values were almost the same as values measured with the fasting treatment. Elevated D-BHB concentrations returned to pre ketogenic diet values 3 hours after refeeding with the control diet. Corresponding to fasting and changes in D-BHB concentration, mast cell degranulation to compound 48/80 were significantly inhibited after 15 and 24 hours after food restriction and retuned to before ketogenic diet ratio 3 hours after re-feeding (Figure 4 middle right).For intradermal infusion of D-BHB, significant increases in serum and intraperitoneal D-BHB concentrations were observed immediately after infusion, which then gradually decreased (Figures 5 lower left and lower center). Serum D-BHB returned to initial levels after 3 hours. Intraperitoneal D-BHB concentrations remained at approximately half the value of the maximal concentrations after 3 hours infusion. Mast cell degranulation was significantly inhibited after D-BHB infusion and the inhibition rate plateaued 1 and 2 hours after infusion. Inhibitory effect was diminished 3 hours after infusion (Figure 5 lower right).

Changes in [Ca²⁺]_i responses to compound 48/80 were analyzed in fura 2-loaded RPMC derived from the ketogenic diet, fasting, and in vitro exposure to 100 mM D-BHB.

Compound 48/80 evoked increases in [Ca²⁺]_i upon a 240 seconds application and these responses were reversible. Compare to RPMC The [Ca²⁺]_i response to compound 48/80 was lower in RPMC derived from the ketogenic diet and fasting group than RPMC derived from free access to the control diet. The amplitude of the [Ca²⁺]_i increase in fasting was similar to that obtained with the ketogenic diet (Figure 6A). The peak increase in [Ca²⁺]_i from baseline (Δ[Ca²⁺]i) was significantly lower in the ketogenic diet and fasting group than that in the control diet group (Figure 6B).

To confirm the effect of D-BHB, we incubated D-BHB with isolated mast cell in vitro. D-BHB inhibited mast cell degranulation in a dose dependent manner (Figure 6C). Significant inhibition was observed at more than 10 mM (Figures 6D and E). In addition, pretreatment with D-BHB decreased compound 48/80 evoked [Ca²⁺]_i and Δ[Ca²⁺]_i was significantly suppressed (Figure 6F).