Evaluation of Reduction of Aflatoxins by Near Infrared Spectrometric Sorting In the Highly Aflatoxin-Contaminated Peanut Lots

Takahashi H

Published on: 2022-10-11

Abstract

This study evaluated that performance of the NIR spectrometric soring (Q-sorter) in naturally and highly AF-contaminated peanut lots in a commercial scale. The sorting clearly showed that AFs could be decreased effectively to the levels of less than 10 µg/kg even though the lot was extremely highly contaminated with more than 700 µg/kg of AFs by rejecting the deteriorated kennels including the internally molded kernels. The NIR sorting can contribute enhancement of the efficient utilization and recondioning of AF-contaminated lots in peanuts industry, and also mitigation of health risk and economic impacts.

Keywords

Spectrometric

Introduction

Aflatoxin (AF) are potent carcinogenic mycotoxins produced by the fungi such as Aspergillus flavus (Link) and Aspergillus parasiticus (Speare). The mycotoxin contamination in agricultural products is a serious global problem concerning human health and economic impact [1]. Numerous strategies and managements have been applied to reduce the entry of AFs to the foods and feeds chains from pre-harvest and post-harvest stages including biocontrol using atoxigenic strains of A. flavus [2-3]. Studies have found that AF contamination in peanuts is not uniform, and a few highly contaminated kernels distributed among a large number of uncontaminated ones [4-5]. Therefore, sorting is useful in the final stage of the processing to reduce the levels of AFs if the contaminated kernels are removed efficiently. Common sorting methods including hand picking and electric color-sorting can mitigate effectively the levels of AFs [6-7]. However, the report showed that these sorting, which is principally dependent on appearance (such as color sorting and handpicking), is effective but not complete, and the efficiency of removal is highly variable [7]. These studies have also demonstrated that a few apparently sound and mature kernels contained high levels of AFs [6], and moreover, a positive relation was observed between AF contents and occurrence of kernels, specifically, “a lumen” or a hollow space inside the kernel when examined internally [7]. These results indicated that sorting methods that depended on appearance and applied reflected light were often unable to eliminate completely the contaminated kernels because they were apparently healthy, internally molded kernels with AF-producing fungi often occurred in AF-contaminated lots.

Spectrometric methods developed as rapid and non-destructive tools to identify AF-contaminated kernels and detect AFs in certain crops [8-9]. A previous report [10] showed that the ratios of transmittance energies at 700 -1100 nm, near infrared (NIR), were different between normal kernels and moldy kernels even though they were internally molded. This NIR spectrometry sorting could probably detect changes which were caused by fungal digestions of cell components such as triglycerides and carbohydrates while it was unable to discover AFs directly in the contaminated kernels. We evaluated performance of the sorter that was developed as a Q-sorter (ANZAI MANUFACTURING Co., LTD, Chiba Japan) in naturally AF-contaminated peanut lots [11].

Peanut sample: The peanuts sample (shelled nuts, China product) was extremely contaminated with AFs in the range of 257.4 - 1146.2 µg/kg and 706.6 µg/kg in average.

Sorter and sorting: Q-sorter Leo-300 MK with belt type carrier system can be processed maximally 1,500 kg /hr. The peanut sample was sorted three times. The passed fraction after 1^st sorting was re-segregated as 2^nd sorting, similarly, the passed fraction taken from the 2^nd sorting was re-segregated as 3^rd sorting.

Visual observation and AFs assay in individual rejected kernels: 300 kernels of peanuts sample (294.9 g) were randomly taken from the rejected fraction (R-3) after the 3^rd sorting and then roughly classified into 5 groups depending on the appearance after the visual inspection; 1) apparently healthy but moldy with often discoloration 2) discolored but without visible mold growth 3) insect-damaged 4) dark plumule 5) apparently healthy. They were split off into two halves of the cotyledon by hands. Then, they were inspected visually presence or absence of the fungal growth and discoloration inside of the kernel by a low magnified microscope.

Results

470 kg of the shelled kernels sample was segregated partitioned into 431 kg of passed fraction (P-1, 91.7 % yield) and 39 kg (R-1, 8.3%, rejected ratio) of the rejected fraction. 706.6 µg/kg of the initial concentration of total AFs in average decreased to 319.8 µg/kg by the first sorting. Decrement of the AF was 54.7% (Table 1). Then, 400 kg of P-1 sample was partitioned to 385kg (96.3% yield) of P-2 and 15kg of R-2 (rejected ratio: 3.8%) by the 2nd sorting. AF contents of the passed fraction became lower to 182.1 µg/kg in average. Decrement of AFs by the subsequent sorting was 43.1 %. Further, 297 kg of P-2 sample was re-segregated to 251 kg of P-3 (83.6% yield) and 46 kg (removed ratio: 16.4%) of R-3. Decrement of AF by the last sorting rose up to 97.4%. The overall decrement of AFs through the tree times by the sorting was 97.4%. Overall weight percent of the rejected categories calculating from cumulative weight (100 kg) was 28.5%.

Visual Observation and AFs Assay

300 kernels of the rejected peanuts were classified as showed in Table 1. Fungal mycelium were observed in eleven kernels of moldy group. AFs were only found in seven kernels of moldy group (7/11), especially in the kernels with abundant mycelial growth and obvious discoloration. AFs were not detected in any other groups of the kernels examined including the dark plumule and apparently healthy groups. These results demonstrated that the spectrometric sorting surely could be eliminated AF-contaminated kernels to the rejected fraction. The results also clearly showed that AFs fund in only in the kernels with the discoloration.

Table 1: Visual inspection and classification of 300 kernels in the rejected fraction (R-3) into the groups segregated by the NIR spectrometric sorter (Leo-300MK) in the extremely highly aflatoxin-contaminated sample.

Fraction	No. of kernels before opening (Weight, g)	Weight percent of the divided group
*Inspected^a**		(No. of kernels)
		Moldy	Discolored	Insect damaged	Dark	Apparently healthy
		Moldy	Discolored	Insect damaged	plumule	Apparently healthy
R - 3	300	11	9 (3.0)	0	22	258
R - 3	(294.9)	(3.7)	9 (3.0)	(0.0)	(7.3)	(86.0)
^*aSee Table 2. (Okano, [11])

Discussion

This study clearly showed that both types of the NIR sorters, Q-sorter Leo-300MK, could eliminate more than 90% of AFs contained in the original sample when approximately. 30% of peanuts were taken into the rejected category by three times of sorting even though the sample were highly AF-contaminated as showed in Tables 2. To get proper AF decrement selection of suitable operating condition was the key because AF decrement changed widely with the removal ratio. Information such as AF content and records of the sample before and after harvest must be very important keys for determination of the condition.

Table 2: Reduction of AFs in the extremely highly AF-contaminated sample by the high performance NIR spectrometric sorter (Leo-300 MK).

Sorting	Sample examined			Segregated fraction after sorting
	Weight (kg)	*AFs con.^a (µg/kg)**		Passed				Rejected
		Range	*Avg.^b**	Category Weight (kg)	AFs conc. (µg/kg)		*AF decrement^c**	Weight (kg)	AFs conc. (µg/kg, avg.)
		Range	*Avg.^b**	Yield (%)	Range	Avg.	(%)	Ratio (%)	AFs conc. (µg/kg, avg.)
1^st	470	257.4 - 1146.2	706.6	P - 1 431 (91.7)	24.0 – 700.3	319.8	54.7	R - 1 39.0	6478.8
1^st	470	257.4 - 1146.2	706.6	P - 1 431 (91.7)	24.0 – 700.3	319.8	54.7	-8.3	6478.8
2^nd	400	24	319.8	P -2 385 (96.3)	11.4	182.1	43.1	R - 2 15.0	3031.2
2^nd	400	-700.3	319.8	P -2 385 (96.3)	-285.3	182.1	43.1	-3.8	3031.2
3^rd	297	11.4	182.1	P- 3 251 (83.6)	1.2	4.7	97.4	R - 3 46.0	70.8
		-285.3			-			-16.4
					7.8

*^aAFs concentration. *^bIn average. N=4. *^cAF decrement = (1-accepted category AFs conc. /AFs conc. before the sorting) × 100. (Okano, [11])

The visual inspection of the kernels after splitting (Tables 3) indicated that the discolored kernels were also likely to eliminate to the rejected fraction by the sorting along with the moldy kernels. They also showed that the internally molded kernels, which were commonly with discoloration, certainly eliminated to the rejected fraction after the sorting. Discoloration in the peanut kernel might be caused by fungal digestion of cell components, especially in lipids, which were richer in peanut contained nearly 50% in the kernel. The main components in the lipids are triglycerides that stored in the form of small droplets called oil bodies. They distributed throughout the whole seed, which were surrounded with ultrathin (2-3.5 nm in width) biological membrane [12]. Therefore, the bodies were likely to decompose by the phospholipase secreted by invaded fungal mycelium. The lipolytic enzyme was reported to be higher potent in Aspergillus spp. [13], especially A. flavus and related species [14]. The bodies also likely to break by physical damage including by insect biting, from which the fungi could invade with ease. The released fatty acids oozed to cells of the peanut tissues, and then, which resulted in the brown discoloration of the kernels. A previous report [9] demonstrated that there was a linear relationship between the transmittance ratio (700/1100 nm) and the degree of the hydrolysis of triglycerides. Therefore, molded kernels would be likely to fall in the rejected fraction by the NIR spectrometric sorting. Moreover, in the group of discolored without visible fungal growth, the kernels might be invaded by the fungal mycelium and the lipids were decomposed even though obious fungal growth was inspected yet.

Table 3: AFs concentration in individual peanut kernel of the moldy group after the visual inspection in the rejected fraction ^*a sorted by the NIR spectrometry (Leo-300 MK).

Kernel No.	AFs concentration			Degree of discoloration in
	B₁ (µg/kg)	G₁	Total in kernel	*the kernel^c**
	B₁ (µg/kg)	(µg/kg)	(µg/g)
1	N. D.^*b	N. D.	N. D.	++
2	N. D.	N. D.	N. D.	-
3	N. D.	N. D.	N. D.	-
4	N. D.	N. D.	N. D.	++
5	74.3	326.4	4.2	+++
6	310.1	99.5	6	++
7	495.3	N. D.	6	+++
8	18.1	N. D.	0.2	+
9	0.4	N. D.	N. D.	+++
10	0.2	N. D.	N. D.	+++
11	0.2	N. D.	N. D.	-

^*aSee Table 1. ^*bNot detected. ^*cDegree of discoloration was symbolized as, -; scarcely, +; a little, ++; partly, +++; almost entirely. (Okano, [11])

The AF assay in the individual kernels in the last rejected category of the highly contaminated lot showed that AF-contaminated kernels were concentrated in the internally molded group but not found in any other divided groups in this study (Table 3). Numerous reports showed that AF-contaminated kernels were commonly present in the discolored, insect-damage and/or shriveled kernels [4,7]. This this study, the apparently discolored kernels with highly levels of AFs might be eliminated to the rejected fraction before the last sorting because the severely deteriorated kernels might be eliminated easier by the preceded sorting depending on the appearance as well as insect-damaged kernels. These results from the individual examinations showed that the spectrometric sorting surely could be segregated the internally moldy kernels and there was a relationship between removal of the internally molded kernels and reduction of the levels of AF in the accepted fractions.

The NIR sorting clearly showed that the NIR spectrometric sorting can contribute enhancement of the efficient utilization and recondioning of AF-contaminated lots in peanuts industry, and also mitigation of health risk and economic impacts through the reduction of AF contamination levels.

Journal of Biomedical Science and Research