Population Explosions of Tiger Moth Lead to Lepidopterism Mimicking Infectious Fever Outbreaks

Lepidopterism is a disease caused by the urticating scales and toxic fluids of adult moths, butterflies or its caterpillars. The resulting cutaneous eruptions and systemic problems progress to clinical complications sometimes leading to death. High incidence of fever epidemics were associated with massive outbreaks of tiger moth Asota caricae adult populations during monsoon in Kerala, India. A significant number of monsoon related fever characteristic to lepidopterism was erroneously treated as infectious fevers due to lookalike symptoms. To diagnose tiger moth lepidopterism, we conducted immunoblots for tiger moth specific IgE in fever patients’ sera. We selected a cohort of patients (n = 155) with hallmark symptoms of infectious fevers but were tested negative to infectious fevers. In these cases, the total IgE was elevated and was detected positive (78.6%) for tiger moth specific IgE allergens. Chemical characterization of caterpillar and adult moth fluids was performed by HPLC and GC-MS analysis and structural identification of moth scales was performed by SEM analysis. The body fluids and chitinous scales were found to be highly toxic and inflammatory in nature. To replicate the disease in experimental model, wistar rats were exposed to live tiger moths in a dose dependant manner and observed similar clinico-pathological complications reported during the fever epidemics. Further, to link larval abundance and fever epidemics we conducted cointegration test for the period 2009 to 2012 and physical presence of the tiger moths were found to be cointegrated with fever epidemics. In conclusion, our experiments demonstrate that inhalation of aerosols containing tiger moth fluids, scales and hairs cause systemic reactions that can be fatal to human. All these evidences points to the possible involvement of tiger moth disease as a major cause to the massive and fatal fever epidemics observed in Kerala.


Introduction
Recent widespread and recurrent fever epidemics that occurred in tropical countries including India have experienced high morbidity and mortality [1,2]. Kerala, the south-western state of India faced disastrous fever outbreaks, where more than 23.4 million people (S1 Fig) [3] were affected leading to approximately 2,250 deaths since the year 2006. Despite having aggressive public health-care practices, recent occurrences of epidemic pose a great challenge to the extensive health programme in the State. The advent of fever epidemics ravaged all the fourteen districts of the State and has drawn worldwide attention due to its explosive nature [1,[4][5][6][7]. In the midst of epidemics, clinicians in our hospital encountered a plethora of weird symptoms and were puzzled as the fever showed hallmark symptoms of chikungunya and at times of dengue.
The tests, however, were negative in a number of cases.
Multitude of unusual symptoms consisting cutaneous signs like maculopapular or petechial rash, soft tissue necrosis with extensive sloughing and hyperpigmentation to systemic complaints including fever with chills, headache, nausea, vomiting and diarrhea, sore throat, ecchymosis, large hives, edema, crippling arthralgia and arthritis, conjunctivitis, pharyngitis, respiratory distress, decrease in platelet counts, multi-organ disorders and reduced electrolyte levels have been observed (S1 Table) [4,5,7]. More seriously, incidence of meconium stained amniotic fluid and meconium aspiration syndrome associated with fever was observed in pregnant victims [6].
To our surprise, symptoms of different infectious diseases originate simultaneously during the epidemics and disappear altogether during decline phase of the epidemics. Further, a significant number of cases were diagnosed negative for almost all known infectious fever diseases (S2 Table), and were classified as suspected cases (S1 Fig) [3], leading to the assumption that the diseases are emanating from a unique source. Here, we investigate the involvement of tiger moth Asota caricae Fabricius (Lepidoptera: Noctuidae: Aganainae) induced toxic-allergic disease in fever patients during monsoon that mimics the symptoms of infectious disease. With the onset of monsoon, millions of tiger moth caterpillars infest host plants in the region and disease erupts during the eclosion of adult moths. On the other hand, fever outbreaks decline when the cyclic event moves into the larval and pupa stages. Adult moths are strongly attracted to artificial night lights and once entered in human habitation, they circle around bright lights, extricate scales, hairs and/or release deadly chemicals into the ambient and inhalation of these blends leads to severe forms of fever syndrome in human. In most instances, these symptoms mimic that of infectious fever outbreaks. In general, the pathophysiologic manifestations of toxic moths and/or caterpillar exposures are classified into distinct clinical syndromes that include erucism (reactions due to caterpillars), lepidopterism (cutaneous and systemic signs), pararamose (severe arthralgia and arthritis due to Premolis caterpillars), lonomism (contact with Lonomia caterpillars), dendrolomiasis (exposure to Dendrolimus caterpillars) and ophthalmia nodosa (ocular involvement) [8][9][10][11][12][13][14][15]. Several other unusual clinical presentations were also reported during the epidemic and one of them was meconium stained amniotic fluid and associated meconium aspiration syndrome in pregnant victims [6]. These conditions have a close similarity with equine fetal loss following exposure to eastern tent caterpillars and processionary caterpillars [16,17]. Prevalence of tiger moth lepidopterism in India is not well documented and we suggest that heavy infestations of tiger moth can facilitate huge epidemics that lead to adverse reactions of lepidopterism or other related syndromes.
In this study, we conducted the chemical characterization of tiger moth fluids which are responsible for the immuno-inflammatory response in humans. Further, we detected tiger moth specific IgE antibody in fever patients' sera and estimated the release of proinflammatory cytokine TNF-α in the associated joint disease. In addition, we replicated tiger moth disease in a rat model based on the clinico-pathological complications observed in fever patients with joint ankylosis. Finally, we cointegrated field estimates of tiger moth caterpillar abundance with incidence of epidemics and with meteorological conditions which are considered to be the factors behind the explosion of tiger moths.

Patients' Survey
Massive fever epidemics occurred in Kerala in the year 2007 and witnessed unusual clinicopathological complications among fever affected patients where most persisting was the fever with joint pain. We conducted a follow-up household survey in the year 2008 among these patients by giving an initial oral description about the fever epidemics and obtained consent for participating in the survey. Each patient was requested to fill a questionnaire with their name and address, age, weight, symptoms of the disease, disabilities during convalescence, prolonged effect of the disease, treatment taken and their perception about the cause of the disease.

Moth Census, Fever Data and Environmental Data
We conducted a survey in Kerala (8°36' to 12°30' N and 74°59' to 77°03' E, altitude up to 770 m above sea level) during July to September 2009 to understand the occurrence, geographical distributions and population sizes of tiger moth in the State. For that we collected data on the number of pioneer figs infested with tiger moth caterpillars and the number of caterpillars per plant from randomly positioned three transects (200 x 2 m) at 10 km intervals. Host plants were represented by three tier habits herbs, shrubs and trees in the region and the abundance of caterpillars was measured on herbs, shrubs and trees separately. In this study, we considered main branches as shoot in herbs and side branches on the main stem as shoots in shrubs and trees. We observed that caterpillars are equally distributed on each shoot of the plant and size of the population was measured in terms of caterpillars per shoot (measured on randomly selected 5 to 25 shoots depending on the size of the plant) and then extrapolated to per plant and finally to the study area. No specific permissions were required for conducting these activities at different locations in Kerala because the invasive host plants were available on both sides of the public roads including national highways, state highways and even small pocket roads. The larvae forage on the leaves when atmospheric temperature drops, so the censuses were carried out in the early morning (between 05:00 and 08:00 h) and late evening hours (18:00 to 24:00 h) and the sampling extended to 06:00 to 11:00 h and 14:00 to 22:00 h on rainy days. We selected four districts Kozhikode, Idukki, Kottayam and Pathanamthitta (represent major stations A-D respectively in Fig 1B) where huge fever outbreaks reported in the year 2007 to study the moth population characteristics and its relation with fever outbreaks. At each station we selected substations (A-9, B-15, C-12 and D-15 substations) and from each substation three transects (each transect 200 x 2 m) at 10 km intervals were marked and data on number of host plants, egg masses, caterpillars, pupae and parasitation were recorded monthly from 2009 to 2012. Size of the population was measured in terms of caterpillars per shoot and then extrapolated to per plant and finally to the study area. The observed larval mortality (due to parasitation and heavy rainfall) was deducted from larval abundance values. The mortality at pupal stages was not considered as it was observed to be very negligible (3.6%, n = 852). Geographic positions of the study area (S1 Dataset) were marked using GPSMAP 76CSx, Garmin and the dataset was exported to geographical information software (ArcGIS 10) to display geographical distributions and population sizes of the moth. Meteorological data on relative humidity, temperature, and rainfall were obtained from Indian Space Research Organization, Centre for Water Resources Development and Management and India Meteorological Department, India. Monthly fever data was collected from State Disease Control and Monitoring Cell and Directorate of Health Services (DHS), Kerala, India. Source of daily fever death reports were taken from DHS website and Malayala Manorama online edition. Geographical range of A. caricae habitats includes Asia to Pacific Islands and the positions were marked using ArcGIS 10 software.

Moth Rearing
Egg masses, larvae and pupae were collected from the field and reared in the laboratory under controlled conditions. We further confirm that the field studies did not involve any endangered or protected plant or animal species. Body fluids collected separately from male and female moths in sterile containers immediately after the emergence and stored at -20°C until use. Secretions from fourth, fifth and late fifth instars were collected from the prothorax region and stored immediately at -20°C until use. Mating experiments were conducted in special containers supplied with diluted honey in water (1:1) as food. Precautionary steps were taken while handling adult moths using protective cloths, gloves and special face masks.

HPLC and GC-MS Analysis
High performance liquid chromatography of moth fluids was performed and the purified fraction was subjected to Gas Chromatograph-Mass Spectrometry (GC-MS) analysis. The HPLC was Konikrom LC system with UV-VIS and Fluorescence Detectors (Konik HPLC, Konikrom). Separation was carried out at ambient temperature using Vydac C 18 column (250 mm x 4.6 mm). The analysis was performed by gradient elution at a flow rate of 1.1 mL/min. The mobile phase composition consisting of Buffer A: 95% Water + 5% Acetonitrile (Merck), Buffer B: Acetonitrile containing 0.01% TFA (Sigma-Aldrich). The detector was set at 280 and 248 nm and peak areas were integrated automatically using Konikrom Plus software programme. For GC-MS, Konik HRGC gas chromatograph linked to MS system equipped with a fused silica column (25 m x 0.25 mm i.d) was used (Konik HRGC-4000 with MS Q12 detector, Konikrom). The temperature programming was 50°C for 2 min, 125°C@15°C/min, continued for 2 min, and 10°C/min up to 275°C for 5 min. Ion source temperature and interface temperature were 225°C, and 150°C respectively. Standard comparison of chemical components was done with NIST Mass Spectral Library Version 2.0.
Histamines and imidazole in tiger moths and caterpillar secretions were quantified by HPLC using standard chemicals (Sigma-Aldrich). Separation was carried out using C 18 column (150 mm x 4.5 mm x 5μm). The injection volume was 10 μL and the analysis was performed by gradient elution at a flow rate of 1.5 mL/min. The gradient mobile phase consisting of aqueous solution of ammonium dihydrogen phosphate (0.05 M) and the final composition of ammonium dihydrogen phosphate was 0.0075 M. The column oven temperature was 25°C. The detector was set at 230 nm and peak areas were integrated (1260 Infinity Series, Agilent).
Volatiles in caterpillar and moth secretions were analyzed by GC-MS analysis. Briefly, 10 μL of moth fluid was reconstituted with acetonitrile (1:1 v/v) and 1.0 μL of the sample was injected directly into the GC (7890A, Agilent) linked to MS system (5975C Inert XL MSD, Agilent) equipped with columns DB-5 and DB-5ms, 30 m x 0.25 mm x 0.25 μm. The temperature programming was 80°C for 5 min, 250°C@25°C/min, continued for 2 min. The injector temperature was 240°C and the detector temperature was 250°C respectively. The flow rate was 1.0 μL/ min. Standard comparison of volatiles was done with NIST Mass Spectral Library Version 2.0. The octan-1-ol in caterpillar and moth secretions was quantified using standard octan-1-ol (Merck). Thymoquinone standard was procured from Sigma-Aldrich.

SEM Analysis
Scales from different parts of the moth (2 females and 2 males) were visualized under SEM (Hitachi SU6600). Prior to imaging, the samples were sputtered with gold (Hitachi, E1010).

Peptide Analysis
Peptide mass fingerprinting was performed on MALDI-TOF-MS (UltrafleXtreme, Bruker Daltonics). Moth secretions were dissolved in a mixture of deionized water containing 5% ACN. α-cyano-4-hydroxycinnamic acid (0.1 mmol/L) was used as the matrix. The samples were sandwiched in the matrix on the MALDI plate using ZIP-Tip and exposed to laser source (power 70-80) and the peaks were detected using pre-calibrated detector (calibrated using angiotensin, insulin and BSA).

Estimation of Total IgE in Patients' Sera
Clinical study was approved by the Institutional Ethics Committee of the Malabar Institute of Medical Sciences, Kozhikode, Kerala (IEC/2011/12). Blood samples were collected from fever patients according to Institute Ethics Committee guidelines and all participants gave written, informed consent. Total IgE was estimated (04827031190, Roche) by electrochemiluminescence immunoassay (Cobas e 411 analyzer, Roche).

Western Blot Analysis
For protein isolation, fully emerged moths were ground and extracted with TEMP buffer (pH 8.0) containing 0.2 mM PMSF. The extracts were centrifuged at 12000 rpm for 15 min at 4°C and isolated proteins were kept at -20°C until use. Whole body extracts (50 μg) were resolved on 12% SDS gel, stained with Coomassie Blue and photographed. Unstained gels were transferred (Trans-Blot Turbo, BioRad) on to PVDF membrane (Trans-Blot Turbo transfer pack, BioRad and Immobilon-P, Millipore), blocked with 3% BSA for 1 hour, blotted with patients' sera (1:1 v/v in 1% BSA) overnight at 4°C, then blotted with monoclonal anti-human IgE-Fc (HRP) antibody 1:1000 dilution (ab99836, Abcam) at room temperature for 2 hours, followed by color development with diaminobenzidine (Sigma-Aldrich) and protein sizes were determined by standard markers (Promega and BioRad). Specific IgE inhibition was performed by pre-incubating patients' sera (1:5 v/v in 3% BSA diluted in TBS-T) with 10-100 μg of moth extract for 16 h at 4°C. The preincubated mixture was then added to the membrane containing separated moth proteins and blotting was performed. Immunoblot experiments were also performed by substituting fever patients' sera with healthy control, IgE hypersensitive to other allergens (atopic dermatitis, dust and pollen) and fever control (IgE < 200 IU/mL) sera. Immunoblots with fever patients' sera were performed on unrelated heterologous allergen extracts of cockroach and teak defoliator moth to check the specificity of the assay. For β-actin loading control, tiger moth total protein was separated on PAGE and transferred to PVDF membrane. After blocking with 3% BSA, the membrane was incubated with monoclonal anti-β-actin antibody (A2228, Sigma-Aldrich) overnight at 4°C. This was followed by blotting with secondary anti-mouse IgG AP-linked antibody (#7056, Cell Signaling) at room temperature for 3 hours. Finally the membrane was developed by BCIP/NBT (Sigma-Aldrich) method.

Protein Sequencing
For sequencing, protein bands on the gel stained with Simply Blue safe stain (Invitrogen) were sliced and dehydrated with 2:1 mixture of acetonitrile (ACN): 50 mM ammonium bicarbonate (Sigma-Aldrich) and washed with 25 mM ammonium bicarbonate. Reduction and alkylation were done with 10 mM dithiothreitol (Sigma-Aldrich) and 100 mM idoacetamide (Sigma-Aldrich) respectively. Digestion was completed with trypsin (sequencing grade, Promega). Finally, extractions were carried out by increasing grade of extraction buffer having 50, 60 and 80% ACN containing 0.1% trifluroacetic acid (Sigma-Aldrich). After purification using ZIP--Tip, the solutions were loaded into the column (injection volume-10.0 μL, cap pump flow rate-2.00 μL/min, nano pump flow rate-0.30 μL/min) on HR-LCMS (1290 infinity UHPLC and Nano HPLC with 6550 iFunnel Q-TOF, Agilent). Sophisticated Analytical Instrument Facility operated by Indian Institute of Technology, Mumbai, India conducted the protein sequencing analyses.

ELISA
Human TNF-α ELISA was estimated according to manufacturer's protocol (KHC3011, Invitrogen Life Technologies). Briefly, added 50 μL of incubation buffer to the pre-designated wells. Then added 100 μL of standards, control and patients serum samples into the microtiter wells and incubated for two hours at room temperature. After washing four times, added 100 μL of biotinylated anti-TNF-α (Biotin conjugate) solution into each well and incubated for one hour. After washing, 100 μL of streptavidin-HRP working solution (4×) was added and incubated for thirty minutes. After washing, 100 μL of stabilized chromogen was added to each well and incubated for thirty minutes in the dark. After incubation, samples were added with 100 μL of stop solution and recorded the absorbance at 450 nm (iMark microplate reader, BioRad). Calculation was done using MasterPlex Reader Fit software (Hitachi).

Animal Experiments
All animals were cared for in strict guidelines with Committee for the Purpose of Control and Supervision of Experiments on Animals (CPCSEA). The study was sanctioned by the Institute Animal Ethics Committee of the MIMS Research Foundation, Kozhikode, Kerala (MRF/IAEC/ 2012/#04). Male and female rats of Wistar and Sprague-Dawley strains (194.2 ± 27.4 gms, mean ± s.d, n = 48) were procured from Sri Chitra Tirunal Institute for Medical Sciences and Technology (SCTIMST), Thiruvananthapuram and College of Veterinary and Animal Sciences, Trissur, India. All animals were maintained at a controlled temperature of 26-28°C with a 12 h light:12 h dark cycle and fed with standard pellet diet and water ad libitum. Rats were challenged to live moths overnight in filtered cages provided with illuminated bright lights. Animals were monitored for the changes in the body including toenail or fingernail breaks, erythematous rash, bleeding eyes, ear or tail rash pedal edema and arthralgia. Rectal temperatures were recorded daily in the morning and evening using a digital thermometer. For platelet functioning study, 40 μL of moth fluid (collected from prothoracic and abdominal points) was administered to each rat using a micropipette (5 μL in alternate nostril at 2 min interval) in a supine position. Peripheral blood smears were prepared to verify the platelet properties prior and during the treatment and at the end of the recovery phase. Animals were sacrificed by terminal bleeding under ketamine (24 mg/kg, i/m) anesthesia. Blood samples were collected separately in K 2 -ETDA coated and serum collection tubes (BD Vacutainer). For serum collection, samples were centrifuged at room temperature and separated sera were stored at -20°C for further analysis. Complete blood count was performed using automated equipment (Coulter LH 780 analyzer, Beckman Coulter). Alanine aminotransferase and creatinine were estimated by photometric-P5P-kinetic and kinetic Jaffe reaction method respectively (Dimension EXL with LM, Siemens).
Rat IgE ELISA was performed according to manufacturer's protocol (ab157736, Abcam). Briefly, added 100 μL of each standard, control and toxin exposed rat serum samples into predesignated wells and incubated the microtiter plates at room temperature for sixty minutes. After washing four times, added 100 μL of 1× enzyme-antibody conjugate to each well and incubated for sixty minutes. After washing, added 100 μL of tetramethylbenzidine (TMB) substrate solution and incubated in the dark for ten minutes. Reaction was stopped by adding 100 μL of stop solution and determined the absorbance at 450 nm. Calculation was done using MasterPlex Reader Fit software.

Tissue Processing and Staining
Small pieces of lungs, liver and kidneys were fixed in 10% buffered formalin and processed in paraffin (TP1020, Leica). Embedded tissues were sectioned (RM2245, Leica), stained with hematoxylin and eosin and examined pathological changes (double-blinded) under the microscope (ECLIPSE 80i & E200, Nikon). Toluidine blue stain was used for mast cell detection and Leishman-Giemsa was used for megakaryocyte detection in bone marrow and spleen respectively. The images were captured using Nikon DS Filc camera.

Statistical Analysis
Chi-square analysis was performed to compare the symptoms among different age groups in the household survey conducted in fever patients. One-way ANOVA was performed to compare the blood sera parameters of healthy individuals with fever patients. Again, blood sera parameters of animals exposed to live tiger moths were compared with normal controls using one-way ANOVA and all the data are expressed as mean ± S.D. Cointegration and causality tests were carried out using EViews 8 and Gretl softwares to test the relationship of larval abundance with fever epidemics and environmental parameters.

Tiger Moth and Herbivory
The dispersal of tiger moth is extensive from Asia to Pacific Islands [18] (Fig 1A) and we observed synchronous populations of moth in entire Kerala (Fig 1B) that are coinciding with massive fever epidemics (Fig 1C). The moth completes its life cycle in an average of 32.6 days with egg phase (2.7 ± 0.6 days; n = = 23 egg mass (S2A Fig), larval phase (S2B Fig) consist of five instars (12.9 ± 0.7; n = 120), pupae (9.5 ± 0.7; n = 157) and adult (7.5 ± 0.9; n = 136) (Fig  2A). New batches of eggs are laid in 3 days and thus we estimate a minimum generation time Final instar forage on pioneer host leaves is active at night but elusive during day light. Larva is blackish with red prothorax and a central black stripe in which on each segment two black spots arranged longitudinally (Ref. 18). The photographs were originally taken by P J Wills (Fig A & B). (C) Lamellar type represents majority of the scales on the head, thorax, wings and abdomen (magnification 100×) (D) Scales on the female ovipositor were mostly of piliform type (magnification 500×). (E) Defensive fluids released from specific bleeding points. Broad arrow show hairy tufts or "flechettes" (piliform type) and narrow ones denote the toxic secretions from teguments. The photograph was originally taken by Anjana Mohan (Fig E). Adult ecloses during the dusk (18:00 to 19:30 hours) and the flight activity of both the genders are high during scotophase and fall to nil during photophase. Interestingly, female moths outnumber males in the populations (sex ratio 1:2.1, n = 643; χ 2 = 77.2, P < 0.01) and mate multiple times. We observed short flights of moths during twilight followed by small aggregation with progression of nighttime and between 20:00 to 23:30 hours, tiger moths (6.4 ± 3.6 moths/aggregation, n = 92 observations, and range 2-13 moths) are attracted to artificial night lighting and light-coloured walls of human habitation.
Tiger moths render direct human contact in coherence with massive distribution and herbivory of pioneer figs in the region. Ficus represents an important component in terms of insect herbivory where the pioneer species suffer massive caterpillar attack while successional species receive less damage. This strategy is because pioneers tend more towards growth and are poorly defended whereas successional figs are well defended [19,20]. Additionally, volatiles of the pioneers influence herbivory positively [21,22] and caterpillars sequester these vital compounds (S3A and S3B Fig) to adults in blending chemicals that assist concomitant functions of the moth.

Caterpillar Menace
Entry of tiger moth caterpillars, mostly final instars, into human habitation produced severe itching among children and elderly. Print media (Malayala Manorama June 12, 2012) reported large number of hospitalization when larval attack elicited wounds in children and worsened the condition due to severe itching in diabetic patients. We quantified the thick yellow fluid accrued in the caterpillar body estimated as 112.5 ± 8.7 μL/caterpillar (n = 10) and found significant levels of imidazole (> 0.001 mg/μL), diisooctyl phthalate and octan-1-ol (> 0.005 mg/ μL) and at this stage, caterpillar fluids could be passed on to the primary setae which serves as a defensive mechanism (18).

Toxic Fluids and Scales of Tiger Moth
Aposematism or warning coloration, the survival tactic of moths, typically exhibits with combination of brilliant colors communicating a message about its bad taste or toxicity [22,23]. Bestowed with dazzling colors, tiger moths flaunt their marked aposematism and if stimulated, dislodge scales, hairs or atomize body fluids to the ambient. The quantity of moth fluids released from prothoracic and abdominal points is estimated as 0.063 ± 0.07 gms/moth dry wt. (n = 8). We explored the moths' arsenal and found that body fluids consisted of histamines The biogenic amines are vasoactive [24,25], serve multiple functions with a tropism for receptors in bone, joints and cartilage resulting in chronic osteoarthritis, migratory polyarthritis and polychondritis [9]. Imidazole, a heterocyclic compound share 1,3-C3N2 ring system in histidine and histamine, is a diapause terminator [26] act as sturdy skin and eye irritants [27]. Further, octan-1-ol and phthalates could also be potential health hazards to human [28,29].
Once tiger moths are attracted to artificial night lights in human habitation, they shed scales by vigorous wing fanning (S6A Fig), and/or discharge toxic fluids during physical or mechanical disturbances (S6B Fig). Scales are of two types, lamellar and piliform [30], are heavily distributed on the head, thorax, abdomen, wings and genitalia. Scales in various sizes (lamellar types, length 364.1 ± 222.2 μm; maximum width 103.4 ± 41.0 μm, n = 12 and piliform type, length 1125.7 ± 84.4 μm; maximum width 4.3 ± 0.5 μm, n = 12) are identified using scanning electron microscopy (Fig 2C and 2D). Furthermore, we observed that fluids released from specialized pores of prothoracic segments (Fig 2E) are spread over the hairy tufts and occasionally, these hairs are lifted by the long siphon of the proboscis and are dispersed into the air. We reckoned a maximum of twelve female moths attracted to human habitation in single occasion with full of detaching scales and hairs. Inhalation of the toxic components from a single moth is enough to cause the disease but cumulative effect of the toxins could generate severity of the disease, sometimes deadly. A 67-year-old female, one of the investigator's relative, touched a live moth in a failed attempt during collection in the year 2009. She developed symptoms of the epidemic in just 24 hours and was bedridden with fever, intermittent chills and vomiting for the next two days. By day three, she had severe throat discomfort and generalized myalgia but subsided without any specific treatment. She was affected with the disease of similar symptoms in the year 2007 and was hospitalized for 4 days. Investigators (n = 3) were affected with the disease during the collection of live moths that attracted to artificial night lights. All of them developed the symptoms including fever, headache, back pain and generalized myalgia by the next day but the effects were transient.

Heterogeneity of Tiger Moth Specific IgE in Human Sera
In a retrospective study, 83.5 percent of fever patients (n = 3,641) admitted to our hospital suspected with symptoms of infectious fevers through the years from 2009 to 2013 were serological negatives for chikungunya, dengue and leptospirosis. In contrast, the patients had elevated levels of total serum IgE (1,216.3 ± 737.5 IU/mL; n = 155 vs healthy control, 106.2 ± 65.0 IU/ mL; n = 15, P < 0.05). Intriguingly, immunoblots with patients' sera confirmed specific IgEbinding proteins that are congruent to tiger moths (78.6%, n = 61). No bands were found on tiger moth proteins against healthy control (n = 10), fever control (111.1 ± 70.7; n = 10), allergic control sera (886.0 ± 713.0, n = 17) and in specific IgE inhibition blots. Correspondingly, no bands were located on cockroach and Hyblea moth proteins against IgE elevated fever patients' sera (n = 15 each) (Fig 3A-3K).

Long-Lasting Effect of the Disease
To understand the severity of the disease on different age groups, we conducted a survey among patients (n = 862) affected in the year 2007 with mean age of 36 years (range: 3.5-86) responded to the study variables (S1 Table). We observed that respondents with fever (n = 763) and joint pain (n = 824) did experience a greater risk of itch and/or rash (n = 591) with fever preceding the itch and/or rash rather than occurring concurrently. The most common report of the rash was blotchy red areas followed by hive-like bumps, locally named as "tomato fever". Moreover, prolonged physical disabilities estimated to be the highest in elderly ( 60 years, n = 93), lower in adults (30 to 59 years, n = 254) and the lowest in children and young adults (< 30 years, n = 95) with no sex preference.

Up Regulation of Tumor Necrosis Factor-α (TNF-α) in Crippling Arthralgia
In fever patients, the knee joint disease was distressingly persisting and even a slight touch or movement was debilitating. TNF-α concentrations are correlated with the severity of the joint disease, because of the more complex interaction of immune modulators at the synovial membrane [13,36]. Based on these observations, we estimated levels of serum TNF-α concentrations, which showed significant difference between joint disease 287.7 ± 407.1 pg/mL (n = 26) and healthy control groups 17.2 ± 3.5 pg/mL (n = 10, P < 0.05). Joint damage or synovitis begins at the synovial membrane where there is an influx of mononuclear cells which produce inflammatory cytokines such as TNF-α and interleukins in the synovial fluid. These cytokines activate resident synovial cells to produce collagenase that mediate destruction of cartilage leading to the development, progression and severity of the disease [37].
Finger or toenail breaks in experimental rats [38] was fixed as the criteria for defining the mean incubation time of rats subjected to intranasal administration of tiger moth toxin (40 μL). It was estimated to be 8.1 ± 2.2 (n = 10) days. Complete blood count and peripheral blood smear evaluation revealed a significant decline in platelet count (7.7 ± 0.7 x10 5 /μL, n = 10 vs control 10.7 ± 0.8 x10 5 /μL, n = 10, P < 0.05) and the counts elevated during the recovery period (10.3 ± 0.8 x10 5 /μL, n = 10, P < 0.05). Platelet morphology was normal with no platelet clumping indicates platelet activation and that also correlates with mean platelet volume. Regenerative responses to the decrease in the number of platelets and increased number of megakaryocytes in the splenic red pulp and bone marrow [39] were examined (Fig 5A-5D). In experimental rats, decline of platelets was significant but no deaths were recorded due to the toxicity of moth fluids. On an average, the platelet count in rats (10.7 x10 5 /μL) is three times higher than that of humans and the decrease in platelets (3.0 x10 5 /μL) might not have reached up to critical limit for the animal survival, but it can be fatal to human averaging 3.0 x10 5 /μL in the circulation.
In addition, serum alanine aminotransferase elevated to 75.0 ± 13.7 IU/L; n = 10 vs control 55.0 ± 4.2 IU/L (n = 10, P < 0.05) and creatinine to 0.91 ± 0.1 mg/dl; n = 10 vs control 0.74 ± 0.1 mg/dl (n = 10, P < 0.05), indicating hepatic and renal injury. At necropsy (n = 5), lethal effect of the moth fluids leading to multi-organ toxicity was evident in histology findings (Fig 5E-5G). Fig 5E and S8B Fig demonstrate extensive lymphocytic proliferation in the peribronchial region and Fig 5H and 5I illustrate recruitment of mast cells in the bronchioles, both conditions have been reported to release pulmonary histamines [40,41]. Most of the lepidopteran insects possess toxic chemicals such as histamines, and histamine releasing peptides that also facilitate immune-inflammatory responses [9][10][11]34]. Interestingly, we found that progression of immune-inflammatory response mediated by the moth fluids has produced higher levels of rat specific IgE but precisely dependent on the quantity of the toxin exposed. Higher levels of IgE (224.0 ± 212.5 ng/mL, n = 15 vs control 12.3 ± 9.0 ng/mL, n = 10, P < 0.05) was quantified in the sera of rats exposed to moth toxins.
Lepidopteran scales consist of chitin, a polysaccharide biopolymer composed of N-acetyl-β-D-glucosamine, is a strong inducer of inflammatory reactions and potentiator of immune responses in mammals. Importantly, chitin compounds specifically binds to receptors on macrophages and activation process may lead to inflammatory reactions. Chitinase act on the chitin skeleton of the scales, and release fragmented chitin, proteins and other antigenic components. Proliferation of lymphocytes is initiated when these proteins are processed by antigen promoting cells, leading to allergic and other immune responses [42][43][44][45].
Late-phase reactions that follow early-phase reactions are thought to cause recurrent and chronic symptoms of allergic individuals. The release of histamines and other mast cell mediators such as interleukins, tumor necrosis factor-α (TNF-α and proteases contribute to initiate late-phase reaction by recruiting and activating inflammatory mediators, and ongoing dysregulation of such mediators cause organ dysfunction [46]. Experimental animals exposed to tiger moths are observed to have augmented recruitment of mast cells (Fig 5H) and severe perivascular lymphocytic proliferation in the lungs leading to bronchoconstriction (Fig 5E and S8B  Fig) which could be the reason for respiratory distress and pneumonitis in patients observed during the epidemic. In addition, fever patients with joint disease had elevated levels of TNF-α,  emphasize that severe arthralgia and edema with walking debility occur in the late-phase of the disease. Interestingly, a very similar and profound debility of joints with edema is observed in animal models (S1 Movie & S2 Movie) support the prevalence of chronic reactions in tiger moth lepidopterism. Precisely, these tissue damages have led to the development of chronic symptoms and severity of the disease, suggest that mechanisms other than those mediated by IgE are also at play.

Tiger Moth Waves Cointegrate with Epidemics
We observed that the incidence of epidemics may occur either by aggregation of local endemic moth population or by immigration from a place under larval attacks. Thus, metapopulation theory better explains the population dynamics of tiger moth explicitly [47]. Interestingly, the moths' haplotypes represent against vicariance that located in Pacific region at a distance of 5,340 km covering Taiwan!New Guinea!Australia suggesting long-distance dispersal of the species [48]. To look for the magnitude of moth population, we used field estimates of caterpillars on pioneer figs and observed successive generations of moths in the area, pointing to the fact that imidazole play a pivotal role in the termination of diapause entry [26]. Here erratic outbreaks occur from January to April, but aggregate as mild wave in May, followed by huge waves from June to August, which then decline in September through December. Therefore, the outbreak of caterpillars appears to be eruptive type exemplifying a positive density-dependent feedback mechanism [49,50].
Repeatedly, epidemics start three to six days after adult moths' eclosion and turn to huge epidemics when populations overlap. To address the direct relation between moth emergence and fever epidemic, we conducted a time series model of larval population against disease incidence (Fig 1C) using the cointegration analysis. We further integrated meteorological conditions with moth abundance and found the existence of cointegration relation between relative humidity, temperature, rainfall and larval abundance (Table 1) using Augmented Dickey-Fuller test [51,52]. At 0.05 level, the cointegration rank test suggests the existence of at most four cointegrating vectors in the system ( Table 2) by methods of Johansen and Juselius [53]. Further, Granger causality test [54] indicate that there is unidirectional causality that drives larval abundance to fever epidemics (F-Statistic = 15.25, P = 0.00001) and the larval abundance is Granger caused by atmospheric temperature (F-Statistic = 6.54, P = 0.003).
Relative humidity governs the development and survival of insects since the body water content needs to be kept at equilibrium. This is perhaps important for the early instars (S2B Fig) which survive better and develop faster when aggregated, possibly to avoid desiccation [55]. Being ectotherms, insects face greatest extinction risks in the absence of either migration or adaptation (S9 Fig) [56][57][58]. We found that tiger moths have high survival and reproductive

Conclusion
During the recent fever outbreaks in Kerala, a large group of patients had clinical symptoms similar to chikungunya and dengue. Interestingly, a significant number of the cases were negative to these infectious fevers and clinicians treated the disease symptomatically. In any of these fever outbreaks, the association between adult tiger moths and the epidemics were not taken in to account and most of these cases were classified as suspected infectious fevers. In clinics, delay in treatment due to a series of investigative procedures had complicated the conditions including platelet drops, respiratory disorders, meconium aspiration syndrome, hepatic and renal failure leading to a number of deaths. In this context, our findings would provide new insights in the development of applied clinical treatment for the fever patients and avert large number of deaths owing to symptomatic treatment that are being currently followed in the State. Improved surveillance networks are required to track the movement of the tiger moths associated to different seasons in other regions where also similar epidemics have been reported. Real-time surveillance and warning system during the build-up phase of caterpillars and pruning host plants by properly trained personnel could manage heavy invasion of the tiger moth populations.
Supporting Information S1 Dataset. GPS coordinates. Geographical coordinates of the study area that represents all the fourteen districts of Kerala where host plants infested with tiger moth caterpillars were recorded from July to September in the year 2009. We also confirm that the GPS coordinates contained with our supporting information files will not lead to the identification of any specific individuals.
(XLS) S2 Dataset. Frequency and duration of itching in rats. Rats exposed to live moths had severe itching all over the body. Frequency and duration of itching was significant in experimental rats exposed to live tiger moths compared to unexposed rats. (XLSX) S1 Fig. Epidemics   Tiger moth caterpillars select cooler microhabitats during warmest hours to avoid maximum operative temperatures. At high temperatures, female tiger moths prefer to lay eggs on host plants that are covered by twining climbers provide microclimate for the growing instars which function as a perfect shelter from soaring temperatures. With the arrival of monsoon, the caterpillars forage on all level of pioneer habits due to the changes in atmospheric temperature, leading to large scale outbreaks of tiger moths. The photograph was taken by P J Wills.
(TIF) S1 Movie. Severe itching. Rat exposed to live moths developed severe itching all over the body. Also observed severe edema on the right leg. (MP4) S2 Movie. Severe arthralgia. Rat exposed to live moths developed severe edema and crippling arthralgia. Debility to walk sustained for 3 to 5 days and recovered by self. Similar presentation was widespread in fever patients during the epidemic. (MP4) S1