Fungal Presence in Selected Tree Nuts and Dried Fruits.

Sixty-four tree nut samples (almonds, pecans, pine nuts, and walnuts) and 50 dried fruit samples (apricots, cranberries, papaya, pineapple, and raisins) were purchased from local supermarkets and analyzed for fungal contamination using conventional culture as well as molecular methods. The results of our study showed that the highest yeast and mold (YM) counts (5.34 log10 CFU g(-1)) were found in walnuts and the lowest in pecans. The most common mold in nuts was Aspergillus niger, relatively low numbers of A. flavus were found across the board, while Penicillium spp. were very common in pine nuts and walnuts. Low levels (2.00-2.84 log10 CFU g(-1)) of yeasts were recovered from only two pine nut samples. Fungal contamination in dried fruits was minimal (ranging from <2.00 to 3.86 log10 CFU g(-1)). The highest fungal levels were present in raisins. All papaya samples and the majority of cranberry, pineapple, and apricot samples were free of live fungi. The most common mold in dried fruits was A. niger followed by Penicillium spp. One apricot sample also contained low levels (2.00 log10 CFU g(-1)) of yeasts.

Introduction

Foods of plant origin, such as tree nuts and fruits, are known to foster the growth of various microorganisms including toxigenic and pathogenic fungal species. Tree nuts in particular have been often reported to contain potentially toxigenic molds such as Aspergillus flavus, A. parasiticus, A. niger, and other species capable of flourishing in this type of substrates. According to past research, peanuts and tree nuts (such as walnuts and pecans) were found to carry such organisms. Bayman et al1 reported that the most common fungi in nuts were Aspergillus, Rhizopus, and Penicillium. A. niger, A. flavus, A. parasiticus, and A. nomius were reported from surveys conducted in the US, Europe, Australia, and South America.2–6 The presence of various aspergilli and penicillia (eg, A. niger, A. flavus, Penicillium crustosum, etc) is of special concern since it has been established that some members of these species are capable of producing mycotoxins. Aspergillus section Flavi (A. flavus, A. parasiticus, and A. nomius) are known to produce the toxic and carcinogenic compounds aflatoxins (AFs) in foods and agricultural products including tree nuts, while black aspergilli (Aspergillus section Nigri) produce 145 secondary metabolites several of which are toxic to humans and animals. More specifically, A. niger could produce ochratoxin A (OTA) and fumonisin B2 (FB2), while the majority of A. carbonarius strains produce OTA. Other possibly toxic metabolites often generated in large amounts by Aspergillus section Nigri are malformins, naphtho-γ-pyrones and bicumarins (kotanins).7 Naturally occurring mycotoxins such as AFs, OTA, and FB2 have also been reported from tree nuts. Low levels (<1.5 ng g−1) of FB2 were detected in a few samples of pick-out almonds.8 AFs (15–25 ng g−1) and zearalenone (ZEA) (125 ng g−1) were found in walnuts.9 Higher levels of AFs (up to 95 ng g−1) were reported in almonds,6,10 while sterigmatocystin (ST) was detected in in-shell pecans.11

Fruits after harvest carry plant pathogenic species from the field, which may or may not survive the drying process. Live potentially toxigenic storage fungi were found in various dried fruits in the past. Tournas (2004, unpublished data) isolated A. niger, A. carbonarius, and A. ochraceus from raisins, while Hocking et al12 reported the presence of Aspergillus section Nigri and A. ochraceus in grapes and grape-derived products such as raisins. The presence of OTA in dried fruits such as currants, raisins, and sultanas has also been reported in the literature.13,14 If the moisture of the dried products is kept at the proper levels, they should not support mold growth. When their water activity (aw), however, surpasses 0.65, they become prone to spoilage by some postharvest fungal pathogens including xerophilic aspergilli and eurotia depending on the aw mark.15 The presence of such species could compromise the quality of the affected products and be the cause of serious health concerns for the consumers. Although reports on fungal profiles of tree nuts and dried fruits exist in the literature, the ongoing changes in fungicide types and levels used may have over time caused changes in mycobiota present on these commodities. Therefore, we deemed it appropriate to revisit the issue and conducted this study in order to evaluate the mycological quality of various tree nuts and dried fruits presently available in the U.S. market.

Materials and Methods

Materials

Dried fruits and tree nuts tested in this study were obtained either in their unopened packages or from bulk jars and trays (1–lb portions) from four supermarket chains (eight stores) in the Washington, D.C. area; each sample was drawn from an individual lot. All samples were kept at room temperature from the time of purchase until commencement of analysis, which was conducted within 48 hours from the time of purchase. Fifty grams from each sample was analyzed for fungal contamination.

Chemicals, reagents, and other supplies

Polymerase chain reaction (PCR) reagents and DNA ladder were purchased from Fisher Scientific. DNA extraction kits were obtained from Norgen Biotek Corporation; primers were purchased from Integrated DNA Technologies (IDT). Agarose gels were obtained from Bio-Rad. All mycological media utilized for the isolation and conventional plating identification of fungal specimens were prepared in-house according to the methods and formulas described by Pitt and Hocking15 and in the Bacteriological Analytical Manual.16

Isolation and quantification of molds and yeasts

Samples were tested as follows: Fifty grams of each product was aseptically transferred into sterile blender jars. Subsequently, they were blended in 450 mL of 0.1% peptone for 45 seconds. Serial dilutions of the homogenates in 0.1% peptone were surface-plated on duplicate DG18 agar (0.1 mL/plate), and the plates were incubated for 5 days at 25°C. Then, colonies were counted, and counts were reported as colony-forming units per gram (CFU g−1).

Speciation of isolated fungal strains

The recovered isolates were purified by reculturing on potato dextrose agar (PDA) (DIFCO) plates and identified to genus or species level using the conventional methods and keys described in Identification of Common Aspergillus Species, A Laboratory Guide to Common Penicillium Species, Fusarium Species: An Illustrated Manual for Identification, and Fungi and Food Spoilage.15,17–19 Isolates that could not be speciated by conventional plating methods, were identified using molecular techniques as described below.

Molecular method—DNA extraction

Fungal DNA was isolated as follows: The fungal strains recovered from tree nuts and dried fruits were grown on solid media at 25°C for 5 days; then a small culture portion from each isolate was transferred to a 15-mL conical tube containing potato dextrose broth and incubated at 30°C for 24 hours. After the incubation period, the cultures were centrifuged for 10 minutes at 10,000 rpm to pellet, and the supernatants were discarded. Subsequently, 10 mL of phosphate buffered saline (PBS) buffer was added to each tube and the pellets were homogenized by vortexing. The tubes were centrifuged again under the same conditions as above, and the supernatants were discarded. Then, 1 mL of PBS buffer was added to each sample and the tubes were vortexed. Subsequently, 50 μl was removed from each tube and transferred to 2.0-mL microcentrifuge tubes to proceed with DNA extraction. DNA extraction was carried out using the Norgen Biotek Fungi/Yeast Genomic DNA Isolation Kit according to manufacturer’s instructions.

Polymerase chain reaction

The primers encoding the β-tubulin gene, Bt2a (5′-GTAACCAAATCGGTGCTGC TTTC-3′) and Bt2b (5′-ACCCTCAGTGTAGTGACCCTTGGC-3′) were used.20 The method described by Asefa et al21 with minor modifications was utilized. PCR was performed in reaction tubes with a final volume of 50 μL consisting of 25 μL Promega GoTaq Hot Start colorless master mix, 21.5 μL Promega nuclease-free water, 1.5 μL DNA template, and 1 μL of each primer. The mixture was spun, and the PCR reaction was run in an Eppendorf 2231 Thermocycler (Eppendorf North America) programmed with the following conditions: Denaturation at 95°C for 10 minutes, 38 cycles of denaturation at 95°C for 30 seconds, annealing at 55°C for 30 seconds, extension at 72°C for 1 minute, and final extension at 72°C for 7 minutes.

Gel electrophoresis

The PCR products were subjected to electrophoresis in 1% agarose gels (Bio-Rad ReadyAgarose ethidium bromide precast gels) to determine the band sizes and to confirm amplification of the DNA. Gels were run using Tris-borate-EDTA (TBE) buffer at 60 V for 70 minutes, and were subsequently visualized and photographed under UV light in an Alpha Imager (Alpha Innotech Corp.).

Sequencing and species identification

Purification and sequencing of the PCR products was performed by MCLAB, and the resulted sequences were trimmed and edited using FinchTV 1.4.0 software. Subsequently, sequence comparisons and speciation were done by utilizing the basic local alignment search tool (BLAST) in Genbank (www.ncbi.nlm.nih.gov/BLAST).

Results

Fungal contamination in tree nuts

In this study, 64 tree nut samples (consisting of pecans, almonds, pine nuts, and walnuts) were tested for the presence and levels of fungal contaminants. All walnut, 91% of pine nut, and 76% of almond samples carried live fungi. Mold incidence was very low (6%) in pecans. The highest yeast and mold (YM) counts (5.34 log10 CFU g−1) were recovered from walnuts, and the lowest levels (2.00 log10 CFU g−1) were found in pecans (Table 1).

Table 1

Fungal contamination levels in various tree nuts and dried fruits.

PRODUCT	NUMBER OF SAMPLES TESTED	YM COUNTS RANGE (log10 CFU g−1)	FREQUENCY*
Tree nuts
Almonds	17	<2.00–4.00	76
Pecans	16	<2.00–2.00	6
Pine nuts	11	<2.00–3.53	91
Walnuts	20	2.65–5.34	100
Dried fruits
Apricots	8	<2.00–2.60	25
Cranberries	10	<2.00–2.30	20
Papaya	10	<2.00	0
Pineapple	12	<2.00–2.00	8
Raisins	10	<2.00–3.86	60

Note:

Frequency is the percentage of contaminated samples.

Abbreviations: YM, yeast and mold; CFU, colony forming units.

A total of 117 fungal isolates including potentially toxigenic Aspergillus, Eurotium, Penicillium, Fusarium, and Alternaria species were recovered during this study (Table 2). The most common organisms were aspergilli (found in 66% of the samples and comprising 46% of the isolates); these organisms were present in all commodities tested. Black aspergilli (A. niger and A. tubingensis) were the most frequent contaminants of tree nuts. A. niger was found in 35% almond, 36% pine nut, and 30% walnut samples at levels sometimes exceeding 3.00 log10 CFU g−1, while A. tubingensis was isolated from almonds and walnuts at high frequency (47% and 60% of the samples, respectively) and sometimes reached levels higher than 4.00 log10 CFU g−1. A. flavus was detected in all tree nut commodities at relatively low levels. This organism was found in 17% of the overall samples, but only one pine nut sample carried >3.00 log10 CFU g−1. The pathogenic species A. fumigatus was also recovered at low levels (2.30 log10 CFU g−1) from one pine nut and one walnut sample. Alternaria species were found in all tree nut types tested in this study except in pecans, but their levels did not reach 3.00 log10 CFU g−1. A number of penicillia including the potentially toxigenic species P. crustosum, P. brevicompactum, and P. polonicum were also isolated from walnuts and/or pine nuts at levels exceeding 3.00 log10 CFU g−1. One walnut sample contained elevated levels (4.26 log10 CFU g−1) of P. brevicompactum, while two other samples of the same commodity carried above 4.00 log10 CFU g−1 of P. crustosum. Eurotia were generally recovered at low levels and frequency during this study. The same was true for Fusarium spp; Fusarium levels reached 3.00 log10 CFU g−1 in only one walnut sample. Other fungal contaminants recovered from tree nuts at low levels and frequencies were Cladosporium spp., Rhizopus spp., and yeasts (Table 2).

Table 2

Fungal species and levels recovered from selected tree nuts.

ORGANISM	CONTAMINATION LEVELS (log10 CFU g−1 – RANGE)
	ALMONDS (n = 17)	PECANS (n = 16)	PINE NUTS (n = 11)	WALNUTS (n = 20)
Alternaria spp.	<2.00–2.84 (12)	ND	<2.00–2.60 (9)	<2.00–2.30 (10)
Aspergillus spp.	<2.00–2.00 (6)	ND	<2.00–2.70 (9)	<2.00–2.48 (5)
Aspergillus flavus	<2.00–2.48 (18)	<2.00–2.00 (6)	<2.00–3.36 (27)	<2.00–2.84 (20)
Aspergillus niger	<2.00–3.00 (35)	ND	<2.00–3.08 (36)	<2.00–3.28 (30)
Aspergillus fumigatus	ND	ND	<2.00–2.30 (9)	<2.00–2.30 (5)
Aspergillus parasiticus	ND	ND	ND	<2.00–2.00 (5)
Aspergillus tubingensis	<2.00–4.00 (47)	ND	ND	<2.00–5.34 (60)
Aspergillus versicolor	<2.00–2.00 (6)	ND	ND	ND
Aureobasidium pullulans	ND	ND	<2.00–2.18 (9)	ND
Cladosporium spp.	<2.00–2.30 (6)	ND	ND	<2.00–2.48 (5)
Diplodia spp.	ND	ND	<2.00–2.30 (9)	ND
Eurotium spp.	ND	ND	<2.00–2.30 (9)	<2.00–2.95 (10)
Eurotium chevalieri	ND	ND	<2.00–2.48 (9)	ND
Fusarium spp.	<2.00–2.18 (12)	ND	ND	<2.00–3.00 (10)
Fusarium musae	ND	ND	ND	<2.00–2.00 (5)
Penicillium spp.	ND	ND	<2.00–3.04 (36)	<2.00–2.95 (15)
Penicillium brevicompactum	ND	ND	ND	<2.00–4.26 (15)
Penicillium crustosum	<2.00–2.90 (6)	ND	ND	<2.00–4.29 (35)
Penicillium echinulatum	ND	ND	<2.00–3.00 (9)	ND
Penicillium georgiense	ND	ND	ND	<2.00–2.00 (5)
Penicillium glabrum	ND	ND	<2.00–2.30 (9)	<2.00–3.18 (5)
Penicillium hispanicum	ND	ND	ND	<2.00–3.00 (5)
Penicillium polonicum	ND	ND	<2.00–3.40 (19)	ND
Penicillium solitum	ND	ND	ND	<2.00–4.04 (10)
Rhizopus spp.	ND	ND	ND	<2.00–3.30 (5)
Yeasts	ND	ND	<2.00–2.84 (18)	ND
No growth	(24)	(94)	(9)	(0)

Note: Numbers in parentheses indicate% samples contaminated with respective organisms.

Abbreviations: CFU, colony forming units; n, number of samples tested; ND, not detected.

Fungal contaminants in dried fruits

The results of the analysis of dry fruits are shown in Tables 1 and 3. Of the 50 dried fruit samples tested, only 25% of the apricots, 20% of the cranberries, 8% of the pineapple, and 60% of the raisins were contaminated with live fungi. No live molds or yeasts were isolated from papaya. The YM levels in the first three commodities were low (2.60 log10 CFU g−1 or lower); raisins generally carried higher YM counts, reaching 3.86 log10 CFU g−1 (Table 1). The main contaminant of raisins was A. niger, found in 50% of the samples, while A. tubingensis was recovered from pineapple and low levels of penicillia were isolated from apricots and cranberries (Table 3).

Table 3

Fungal species and levels recovered from various dried fruits.

ORGANISM	CONTAMINATION LEVELS (log10 CFU g−1 – RANGE)
	APRICOTS (n = 8)	CRANBERRIES (n = 10)	PAPAYA (n = 10)	PINEAPPLE (n = 12)	RAISINS (n = 10)
Aspergillus niger	ND	ND	ND	ND	<2.00–3.86 (50)
Aspergillus tubingensis	ND	ND	ND	<2.00–2.00 (8)	ND
Cladosporium spp.	ND	ND	ND	ND	<2.00–2.30 (10)
Penicillium spp.	<2.0–2.60 (12)	<2.0–2.00 (10)	ND	ND	ND
Penicillium palitans	ND	<2.00–2.30 (10)	ND	ND	ND
Yeasts	<2.00–2.00 (12)	ND	ND	ND	ND
No growth	(75)	(80)	(100)	(92)	(40)

Note: Numbers in parentheses indicate% samples contaminated with respective organisms.

Abbreviations: CFU, colony forming units; n, number of samples tested; ND, not detected.

Discussion

In this study, the mycological quality of various tree nuts and dried fruits sold in U.S. retail was examined. Tree nuts were found to contain sometimes high viable counts of potentially toxigenic molds such as A. niger, A. tubingensis, and Penicillium spp. A. flavus and Fusarium spp. were also encountered but at a lower rate. Past studies have shown similar contamination profiles of these commodities. According to Jimenez et al,10 penicillia, aspergilli, and Rhizopus spp. were the most common molds found in almonds and other nuts. A. flavus, A. niger, as well as Penicillium and Rhizopus spp. were isolated from 76%, 71%, 86% and 50% of almond samples, respectively. Some of these isolates were capable of producing one or more of the mycotoxins: AFs (AFB1, AFB2, AFG1, AFG2), OTA, patulin (PAT), ST, citrinin (CIT), ZEA, penicillic acid, and griseofulvin. Palumbo and O’Keefe,8 after testing inedible pick-out almonds for fungal contamination, found that Aspergillus section Nigri members (eg, A. tubigensis, A. niger, and A. awamori) were frequent contaminants. A. niger and A. awamori comprised 27% of all mold strains recovered. Seventy-two percent of these strains produced FB2 on CY20S agar, and 19% produced OTA. Kenjo et al22 reported the presence of Aspergillus sections Nigri and Flavi (A. niger, A. flavus, A. nomius, A. parasiticus, and A. parvisclerotigenus) as well as Penicillium, Cladosporium, and Rhizopus spp. in commercial almond powder. Total fungal counts ranged between <2.00 and 3.93 log10 CFU g−1. AFB1 and AFB2 were produced by some A. flavus and A. parvisclerotigenus strains, while AFB1, AFB2 AFG1, and AFG2 were elaborated by all A. nomius and A. parasiticus isolates. Bayman et al,1 after testing 1100 tree nut samples (including almonds, pistachios, walnuts, and Brazil nuts) without visible insect damage, also reported that the most common molds in these commodities were Aspergillus, Penicillium, and Rhizopus. Harvest and postharvest practices seemed to affect the nut mycoflora. Rodrigues et al6 tested almonds for the presence of Aspergillus section Flavi molds. These investigators reported that A. parasiticus was the most frequent, comprising 56% of the isolates, followed by A. flavus (36% of isolates) and Aspergillus tamarii (8% of isolates). Twenty-eight percent of A. flavus strains produced AFBs, while 100% of A. parasiticus isolates elaborated AFBs and AFGs.

Our study revealed the presence of the potentially toxigenic molds A. niger, A. flavus, Penicillium spp. and P. polonicum in pine nuts at levels higher than 3.00 log10 CFU g−1. Fungal contamination of pine nuts has also been reported in past studies. Weidenborner23 determined that the dominant fungal contaminants of pine nuts were Cladosporium (37% of the isolates) followed by Phoma (19% of the isolates), and that 16 of the 31 isolated species (including Alternaria alternata, Aspergillus versicolor, A. niger, Eurotium chevalieri, Penicillium aurantiogriseum, P. viridicatum, P. citrinum, P. crustosum, P. puberulum, P. expansum, P. glabrum, and Trichothecium roseum) were potentially toxigenic. Marin et al,24 on the other hand, reported the presence of Fusarium proliferatum in pine nuts. Fifty-four percent of the F. proliferatum strains isolated by these investigators produced fumonisin B1 (FB1) in shelled pine nuts, while 18% elaborated the same toxin in whole pine nuts in a laboratory study. During our study, no live fusaria were found in this commodity.

According to the results of our study, the most frequent molds found in walnuts were penicillia (including P. brevicompactum, P. crustosum P. solitum, P. glabrum, and P. hispanicum), A. tubingensis, A. niger, and A. flavus. Fusarium spp. were isolated from only 3 out of 20 samples. Past research on fungal contamination of walnuts has shown similar microfungi profiles. Abdel-Hafez and Saber9 reported that A. flavus, A. niger, A. fumigatus, Cladosporium spp., Penicillium chrysogenum, and P. oxalicum were the most frequent contaminants, while Fusarium verticillioides, F. equiseti, and F. oxysporum were less common in walnuts in Egypt. Bacchetti and Arp,25 on the other hand, reported the isolation of a P. crustosum strain from moldy walnuts; this isolate was able to produce penitrem A toxin.

In our study, pecans had minimal fungal contamination. This is in contrast with past studies that reported the presence of various mold contaminants in pecans with potential to produce mycotoxins.11,26 Schroeder and Hein11 demonstrated the presence of A. flavus (found in 43% of the samples) and A. glaucus (encountered in 35% of tested samples) in in-shell pecans; these organisms are known to produce ST. Trucksess et al,26 on the other hand, reported that some A. flavus and A. tamarii strains isolated from pecans were capable of producing cyclopiazonic acid (CPA). Propylene oxide (PO) or other fungicidal treatment(s) applied to pecans after harvest may be responsible for the low YM numbers found in our study. Blanchard and Hanlin27 demonstrated that PO application caused the destruction of 80%–92% of surface microorganisms.

Among the dried fruits tested during our experiment, only raisins showed a notable degree of fungal invasion, mainly by A. niger. Past studies also showed similar mold profiles in raisins. Iamanaka et al28 reported that raisins (black and white sultanas) were contaminated with potentially toxigenic molds. The most frequently isolated fungi from that study were A. niger, followed by A. ochraceus and A. carbonarius. The presence of this group of fungi is troublesome because they are capable of producing mycotoxins in vine fruits. In general, members of the A. niger aggregate (eg, A. niger and A. tubingensis) were found to be more prevalent on grapes than A. carbonarius but a much higher percentage of the A. carbonarius strains were OTA producers.29–31 According to Iamanaka et al,28 15% of A. niger, 60% of A. carbonarius, and 87% of A. ochraceus strains isolated from black and white sultanas were able to produce OTA. The latter study also revealed the presence of OTA in 33% of the black sultana samples at levels above 5 ng g−1. OTA-producing A. niger var. niger and A. tubingensis were also isolated from dried vine fruits or grapes by other investigators,3,32–34 while aflatoxin-producing A. flavus was isolated from grapes in Lebanon.35 According to our findings, a small percentage of the dried apricot and cranberry samples contained very low levels of penicillia, while papaya was free of live fungi. Iamanaka et al28 also reported the absence of live molds from some dried fruits including apricots. Scarcity or absence of live fungi from these commodities perhaps was due to a postharvest microbe-destructing treatment.36

Conclusions

From the results of our study and past literature, it is obvious that mold and mycotoxin contamination of tree nuts and raisins is an ongoing problem. Measures to reduce fungal spore numbers carried on the vine fruits and nuts from the orchard as well as adherence to good manufacturing practices (GMPs) during and after harvest and quick drying are essential for the prevention or minimization of mold growth and mycotoxin contamination in the dried products. Additionally, in order to preserve the quality of these commodities, the products should be kept under conditions that ensure maintenance of appropriate moisture levels throughout storage and marketing. Since nuts contain very low levels of soluble carbohydrates, a small increase in moisture content (eg, condensation due to temperature changes during transport and storage) can result in substantial increase of aw. Increase of aw above a certain level will promote the growth of some fungal species, which can cause spoilage and possibly produce mycotoxins.

Low mold contamination in pecans and some dried fruits perhaps reflects the use of some kind of fungicidal treatment after harvest and before storage and marketing (eg, PO or heat application). If such mold-inactivating treatment was applied, any mycotoxins or spoilage byproducts formed before this treatment most likely would still be active and hazardous to human health. Therefore, future studies should be directed toward testing and establishing mycological profiles of these commodities using samples collected prior to application of mold-eradicating processing.