A 56-year-old female lettuce farmer was admitted to our department in May with a 2-week complaint of low-grade fever, worsening cough, and dyspnoea on exertion. She was treated with 200 mg of amikacin and 500 mg of azithromycin by a family physician, but the therapy had no effect on her low-grade fever, cough or dyspnoea on exertion. The patient had been followed for thyroid nodule. She had no history of smoking, alcohol consumption, animal breeding, bird breeding or chemical agent exposure. There was no family history of atopy. She had not experienced episodic wheezing or rhonchi. She had been working as a lettuce farmer for 36 years. The lettuce harvest time was from December to June, with the largest harvest in April.
Her body temperature was 37.9°C, and her percutaneous oxygen saturation was 91% under ambient air conditions. Clubbing was not noted on her fingers. Physical examination of the head, neck, and abdomen was unremarkable. However, auscultation of her chest revealed bilateral wheezing during inspiration. Her arterial blood gas measurements under ambient air conditions were as follows: pH, 7.431; pO2, 65.4 mmHg; pCO2, 36.9 mmHg; HCO3, 24.2 mmol/L. Her leukocyte count was 7,330/μL. The value of C-reactive protein was 3.55 mg/dL (normal range < 0.60 mg/dL).
Her chest X-ray revealed infiltration in the right upper lobe (Fig. 1A). Plain computed tomography (CT) revealed mild infiltration, ground-grass opacities and reticulo-nodular shadows in both of the upper lobes, indicating interstitial pneumonia (Fig. 1B).
Pulmonary function test (PFT) revealed mixed disorder of both obstructive and restrictive type (Fig. 2A), although PFT revealed normal finding during non-harvesting season (Fig. 2B). Intriguingly, the level of fractional exhaled nitric oxide (FeNO) was 42 ppb, suggesting that eosinophilic inflammation was present in the airway. Skin prick test was positive for lettuce (>8 mm wheal Size). Specific IgE test showed multiple allergies for timothy grass, orchard grass, cedar pollen, hinoki cypress pollen, Japanese white birch, cat, wheat, soybeans, peanut, latex, sesame, crab, kiwi fruit, peach and tomato, although specific IgE test for lettuce has not been commercially available. So far, a clear partial cross-reactivity was reported between lettuce and Platanus-pollen extract [9]. On the other hand, specific IgE test was negative for alternaria and candida. Moreover, she did not have antibody for Trichosporon asahii which is the major pathogen causing hypersensitivity pneumonitis (HP) in Japan.
Bronchoalveolar lavage fluid (BALF) obtained from the right upper anterior lobe (40/100 ml) showed elevated total cell counts (19.7×102 /µL), with an increased percentage of eosinophils (8.5%). The ratio of CD4/CD8 lymphocytes was not detected because the percentage of lymphocytes was low. The BALF culture was negative for both bacteria and fungi.
The laboratory test results are shown in table 1. The patient’s Krebs von den Lungen-6 (KL-6) level was 226 U/mL (normal range < 500 U/mL) and the D-glucan value was decreased to <11 pg/mL. These results suggest that interstitial pneumonia and fungal infections are not the cause of pneumonia, respectively. Her serum total immunoglobulin-E (IgE) level was greatly increased (3038.9 IU/mL). After admission to our hospital, her clinical symptoms gradually improved within a week with 0.5 mg/kg prednisolone (PSL). Even after a gradual decrease in the PSL, her dyspnoea did not flare up. However, she had general malaise and a mild dry cough during the lettuce harvesting season without abnormal shadow on Chest X-ray. The 3-year follow-up showed that her total IgE level increased in December, peaked in May, and suddenly decreased in August. This result was consistent with the lettuce harvest season. The leaf harvest period started in December, peaked in May and ended in June.
Immunoblot analysis
A fresh lettuce centre core was cut using a sharp scalpel, and the white juice was spread on a section and collected. This lettuce centre core juice was dissolved in distilled water to prevent hardening (i.e., 20 µL of white juice in 500 µL of distilled water).
The lettuce centre core juice protein (approximately 15 µg of protein) was separated through sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE). Proteins on the gel were stained with Coomassie Brilliant Blue R-350 (GE Healthcare, Chicago, USA) to detect the total protein patterns (Fig. 3A).
Immunoblot analysis was conducted by transferring the SDS-PAGE gel onto an Immobilon-PTM PVDF membrane (Merck Millipore, Burlington, MA) by using a semi-dry blotting method [10]. The membrane was incubated in 10 mM PBS-T (pH 7.5) and 5% skim milk for blocking. The membrane was then incubated overnight at 4oC in diluted serum (20-fold) in the same blocking buffer. After washing the membranes 4 times with PBS-T for 10 min, the bound primary antibodies were detected by using 5000-fold HRP-conjugated goat anti-human IgG mouse-monoclonal antibody (Jackson Immunoesearch Laboratory, West Grove, PA) and an ECL western blotting kit (GE Healthcare, Boston, MA). After washing the membranes 4 times with PBS-T for 10 min, the resultant chemiluminescent signals were detected on X-ray film (Hyperfilm MP, GE Healthcare). The sera from non-atopic healthy volunteers were used as negative controls.
Fig. 3B shows the binding band of the specific IgG from the sera of the patient and a non-atopic healthy volunteer. The 37 kDa band was recognized in only the patient’s serum. Control immunoblot assays with sera from non-atopic patients did not show any IgG-binding bands.