Endophytic Alternaria and Fusarium species associated to potato plants (Solanum tuberosum L.) in Iran and their capability to produce regulated and emerging mycotoxins

Abstract

Endophytic fungi live inside virtually every plant species, without causing any apparent disease or damage to the host. Nevertheless, under particular conditions, mutualistic lifestyle of endophytes may change to pathogenic. In this study, the biodiversity of Alternaria and Fusarium species, the two most abundant endophytic fungi isolated from healthy potato plants in two climatically different regions of Iran, Ardebil in the north-west and Kerman in the south-east, was investigated. Seventy-five Fusarium strains and 83 Alternaria strains were molecularly characterized by multi-locus gene sequencing. Alternaria strains were characterized by the sequences of gpd and caM gene fragments and the phylogenetic tree was resolved in 3 well-separated clades. Seventy-three strains were included in the clade A, referred as Alternaria section, 6 strains were included in clade B, referred as Ulocladioides section, and 4 strains were included in clade C, referred as Infectoriae section. Fusarium strains, identified by sequencing the translation elongation factor 1α (tef1), β-tubulin (tub2) and internal transcribed spacer (ITS) genomic regions, were assigned to 13 species, viz. F. brachygibosum, F. clavum, F. equiseti, F. flocciferum, F. incarnatum, F. nirenbergiae, F. nygamai, F. oxysporum, F. proliferatum, F. redolens, F. sambucinum, F. solani and F. thapsinum. Twenty-six selected strains, representative of F. equiseti, F. nirenbergiae, F. oxysporum, F. nygamai, F. proliferatum, and F. sambucinum, were also tested for production of the mycotoxins deoxynivalenol (DON), nivalenol (NIV), diacetoxyscirpenol (DAS), T-2 toxin (T-2), beauvericin (BEA), enniatins (ENNs), fumonisins (FBs), fusaric acid (FA) and moniliformin (MON). None of the tested strains produced trichothecene toxins (DON, NIV, DAS and T-2). Two out of 2 F. equiseti isolates, 1/6 F. oxysporum, 1/3 F. proliferatum, and 1/9 F. nygamai did not produce any of the tested toxins; the rest of strains produced one or more BEA, ENNs, FBs, FA and MON toxins. The most toxigenic strain, F. nygamai ITEM-19012, produced the highest quantities of FBs (7946, 4693 and 4333 μg/g of B1, B2, and B3 respectively), along with the highest quantities of both BEA (4190 μg/g) and MON (538 μg/g). These findings suggest that contamination of potato tubers with mycotoxins in the field or at post-harvest, due to a change in lifestyle of endophytic microflora, should be carefully considered and furtherly investigated.

1. Introduction

Under natural conditions plants are not individual entities; they are associated with microorganisms to form the plant holobiont [1]. The nature of the interaction of these microorganisms with plants varies from mutualism to parasitism [2]. Endophytic fungi live inside healthy plant tissues without causing any apparent disease symptoms or damage to their hosts. Endophytic fungi have been recovered in all plant tissues, including leaves, stems, roots, flowers and fruits [3], and every plant examined to date was found to harbour at least one species of endophytic fungi. Research on the ecology of endophytic fungi mostly supports their non-pathogenic nature and their capability to enhance biotic and abiotic stress tolerance [4] and improve mineral nutrition in the host plant [5,6], so that endophytes have been proposed as biocontrol agents or beneficial bioinoculants [7]. Nevertheless, conceptual aspects related to the ecological nature of endophytes are still under debate, because some reports have highlighted that, under specific conditions, the mutualistic relationship of endophytes with plants may change to be detrimental [8].

Potato (Solanum tuberosum L.) is the fourth-largest world crop, surpassed in total production only by wheat, rice and corn. It is a rich source of carbohydrates for human nutrition and an important part of the diet for over 1.5 billion people worldwide. Iran is the fourth-largest potato producer in Asia, where the crop is extensively cultivated in the provinces of Ardebil, Hamedan, Esfahan, East Azarbaijan, and South Kerman, with the Ardebil province being the most important production area because of its favourable climate conditions [9]. During a large-scale investigation on endophytic fungi of potato plants in two major potato-producing areas of Iran, viz. Ardebil area (northwestern Iran) and South Kerman area (southeastern Iran), almost 400 endophytic fungal strains were isolated from above-ground and below-ground organs of potato plants, and 22 fungal genera, and 52 fungal species were identified [10]. In both the regions investigated, Alternaria was the most abundant genus in the above-ground plant parts (stems and leaves) and Fusarium in the below-ground plant parts (roots and tubers). Both Alternaria and Fusarium were previously reported as endophytic in potato. O'Callaghan et al. [11] examined the microbial communities of magainin-producing transgenic lines of S. tuberosum in New Zealand, and identified six species of Fusarium and two species of Alternaria. In Germany, Gotz et al. [12] isolated and identified root endophytic fungi from two different potato lines by traditional techniques and cultivation-independent DNA-based methods. They obtained 67 isolates of Fusarium and 73 isolates of Alternaria. Marak and Kayang [13] isolated and identified endophytic fungi associated with potato plants from South-West Garo Hills, Meghalaya, India. In that survey, the species F. oxysporum, F. redolens, F. semitectum, F. solani, F. sporotrichioides, A. alternata, A. brassicicola, and A. solani were identified.

Alternaria species that cause “early blight” and “brown spot” on potato, as well as on tomato, are A. tomatophila E.G. Simmons, A. solani Sorauer, A. alternata (Fr.) Keissl., A. tenuissima (Kunze) Wiltshire, A. infectoria E.G. Simmons and A. arborescens E.G. Simmons [14]. Alternaria solani is the second most devastating foliar pathogen of potato crops after Phytophthora infestans, worldwide [15]. The common symptoms of Alternaria diseases are necrotic lesions on leaves, which are created by the diffusion of fungal toxins [16].

Fusarium dry rot is a postharvest fungal disease affecting potato tubers, which is a cause of up to 60% loss of stored tubers [17]. The disease is caused by several species of Fusarium, mainy F. solani var. coeruleum, F. sambucinum, F. oxysporum, F. avenaceum, and F. culmorum. Some Fusarium species associated with the disease are known to produce mycotoxins that have been implicated in acute toxicoses of humans and domesticated animals [17,18]. In addition, the prolonged exposure to low sub-lethal doses of mycotoxin may result in chronic toxicoses which often are associated with development of cancer. Several Fusarium mycotoxins are regulated in most countries [19]. In addition, concerns about the “emerging mycotoxins” which may co-occur with regulated ones and contribute to the overall health risk of the contaminated commodities, have been growing [19]. Fusarium graminearum, a major mycotoxigenic species that produces the regulated mycotoxins zearalenone and tricothecenes in corn and small grains, was reported as a causal agent of Fusarium dry rot of potato in USA [20]. It has also been shown that this species is able to produce the trichothecene mycotoxins deoxynivalenon (DON) and nivalenol (NIV) in rotten potato tuber tissue [21]. Potato tubers artificially infected with F. sambucinum, another toxigenic species, contained the trichothecene toxin diacetoxyscirpenol (DAS) in concentrations up to 200 μg/tuber [22]. Also, the emerging mycotoxins enniatins were found in potato tuber tissue infected by a complex of six Fusarium strains from different sources [23].

In this article, we present the molecular characterization of endophytic strains of Alternaria spp. and Fusarium spp., isolated from leaf, stem, root, and tuber of potato plants from two geographically diverse potato-producing areas in Iran. We also investigated the mycotoxigenicity of the Fusarium strains isolated from tubers and roots, to assess their potential risk of mycotoxin production.

2. Materials and methods

2.1. Sample collection, isolation, and preservation of the endophytic Alternaria and Fusaria

Eighty mature and disease symptomless potato plants, thirty-five from the Ardebil area (northwestern Iran) and forty-five from the South Kerman area (southeastern Iran) were collected in August 2018 and February 2019. In each inspected field, two potato plants were collected from the opposite sides of each field, and the points of sampling were geolocalized. The visited fields were at least 5 km apart from each other. Samples were kept in paper bags under refrigeration at +4 °C and immediately transferred to the laboratory for further processing. Plant organs were surface sterilizedwithin 72 h from collection, as described by Alijani Mamaghani et al. [10]. Briefly, samples were immersed in 70% (v/v) ethanol for 2 min, then in 5% (w/v) sodium hypochlorite for 5 min, rinsed 3 times in sterile distilled water for at least 5 min, dabbed between sterile tissues and let dry in sterile conditions for at least 15 min. Isolations of endophytic fungi from stems, leaves, roots and tubers were done on antibiotic-supplemented Potato Dextrose Agar (PDA, Merck, Darmstadt, Germany) followed by single germinated-spore isolation. The cultures were preserved on sterile filter paper pieces stored at −20 °C. Pure cultures of the endophytes were deposited in the Mycology Laboratory of the College of Agriculture and Natural Resources, University of Tehran, Iran; representative isolates that were used for molecular identification and analysis of mycotoxins were also cryopreserved at −80 °C in the Agro-Food Microbial Culture Collection of Institute of Sciences of Food Production, CNR, Italy (http://server.ispa.cnr.it/ITEM/Collection/) under an “ITEM” accession number (Table 1, Table 2).

Table 1.

Endophytic Fusarium strains used for molecular characterization, and accession numbers for their gene sequences in NCBI GeneBank database. The strains used for analysis of mycotoxigenicity are in bold.

Origin and Strain No.(a)	ITEM No..(b)	Plant part	Species	Gene Bank Accession number
				tef1	tub2	ITS
Ardebil Province
FU-41	18962	Root	F. clavum	OQ419383	OQ419308	OQ404982
FU-1	18964	Root	F. equiseti	OQ419350	OQ419275	OQ404949
FU-3	18956	Stem	F. equiseti	OQ419372	OQ419297	OQ404971
FU-73	18965	Root	F. equiseti	OQ419402	OQ419327	OQ405001
FU-87	18966	Root	F. equiseti	OQ419414	OQ419339	OQ405013
FU-94	18967	Tuber	F. equiseti	OQ419422	OQ419347	OQ405021
FU-96	18968	Root	F. equiseti	OQ419423	OQ419348	OQ405022
FU-103	18969	Stem	F. equiseti	OQ419352	OQ419277	OQ404951
FU-34	18960	Root	F. flocciferum	OQ419376	OQ419301	OQ404975
FU-84	18972	Root	F. nirenbergiae	OQ419411	OQ419336	OQ405010
FU-86	18973	Root	F. nirenbergiae	OQ419413	OQ419338	OQ405012
FU-97	18974	Root	F. nirenbergiae	OQ419424	OQ419349	OQ405023
FU-2	18941	Tuber	F. oxysporum	OQ419362	OQ419287	OQ404961
FU-35	18946	Tuber	F. oxysporum	OQ419377	OQ419302	OQ404976
FU-66	18986	Tuber	F. oxysporum	OQ419398	OQ419323	OQ404997
FU-70	18987	Root	F. oxysporum	OQ419399	OQ419324	OQ404998
FU-92	18994	Leaf	F. proliferatum	OQ419420	OQ419345	OQ405019
FU-36	18990	Root	F. redolens	OQ419378	OQ419303	OQ404977
FU-38	18991	Root	F. redolens	OQ419380	OQ419305	OQ404979
FU-40	18992	Root	F. redolens	OQ419382	OQ419307	OQ404981
FU-59	18997	Tuber	F. solani	OQ419391	OQ419316	OQ404990
FU-60	18998	Tuber	F. solani	OQ419393	OQ419318	OQ404992
FU-63	19000	Root	F. solani	OQ419396	OQ419321	OQ404995
FU-64	–	Root	F. solani	OQ419397	OQ419322	OQ404996
FU-72	–	Root	F. solani	OQ419401	OQ419326	OQ405000
FU-85	19003	Stem	F. solani	OQ419412	OQ419337	OQ405011
Kerman Province
FU-14	18957	Tuber	F. brachygibbosum	OQ419356	OQ419281	OQ404955
FU-81	18961	Root	F. brachygibbosum	OQ419408	OQ419333	OQ405007
FU-30	18970	Root	F. equiseti	OQ419373	OQ419298	OQ404972
FU-37	18971	Root	F. equiseti	OQ419379	OQ419304	OQ404978
FU-83	18963	Root	F. incarnatum	OQ419410	OQ419335	OQ405009
FU-20	18975	Tuber	F. nirenbergiae	OQ419363	OQ419288	OQ404962
FU-21	18976	Root	F. nirenbergiae	OQ419364	OQ419289	OQ404963
FU-24	18977	Tuber	F. nirenbergiae	OQ419367	OQ419292	OQ404966
FU-25	18945	Tuber	F. nirenbergiae	OQ419368	OQ419293	OQ404967
FU-29	18978	Tuber	F. nirenbergiae	OQ419371	OQ419296	OQ404970
FU-33	18979	Root	F. nirenbergiae	OQ419375	OQ419300	OQ404974
FU-78	18980	Root	F. nirenbergiae	OQ419405	OQ419330	OQ405004
FU-88	18981	Tuber	F. nirenbergiae	OQ419415	OQ419340	OQ405014
FU-89	18982	Root	F. nirenbergiae	OQ419416	OQ419341	OQ405015
FU-91	18983	Tuber	F. nirenbergiae	OQ419419	OQ419344	OQ405018
FU-4	18947	Tuber	F. nygamai	OQ419381	OQ419306	OQ404980
FU-5	18948	Tuber	F. nygamai	OQ419388	OQ419313	OQ404987
FU-6	18984	Root	F. nygamai	OQ419392	OQ419317	OQ404991
FU-10	18949	Root	F. nygamai	OQ419351	OQ419276	OQ404950
FU-11	18985	Tuber	F. nygamai	OQ419353	OQ419278	OQ404952
FU-12	18950	Tuber	F. nygamai	OQ419354	OQ419279	OQ404953
FU-15	18951	Tuber	F. nygamai	OQ419357	OQ419282	OQ404956
FU-23	18952	Tuber	F. nygamai	OQ419366	OQ419291	OQ404965
FU-26	18953	Root	F. nygamai	OQ419369	OQ419294	OQ404968
FU-27	19005	Tuber	F. nygamai	OQ419370	OQ419295	OQ404969
FU-31	19006	Root	F. nygamai	OQ419374	OQ419299	OQ404973
FU-43	19007	Root	F. nygamai	OQ419385	OQ419310	OQ404984
FU-44	19008	Root	F. nygamai	OQ419386	OQ419311	OQ404985
FU-48	19009	Root	F. nygamai	OQ419387	OQ419312	OQ404986
FU-54	–	Tuber	F. nygamai	OQ419389	OQ419314	OQ404988
FU-71	19011	Root	F. nygamai	OQ419400	OQ419325	OQ404999
FU-77	19012	Root	F. nygamai	OQ419404	OQ419329	OQ405003
FU-79	19013	Root	F. nygamai	OQ419406	OQ419331	OQ405005
FU-90	19014	Tuber	F. nygamai	OQ419418	OQ419343	OQ405017
FU-8	18942	Tuber	F. oxysporum	OQ419407	OQ419332	OQ405006
FU-9	18943	Tuber	F. oxysporum	OQ419417	OQ419342	OQ405016
FU-18	18944	Stem	F. oxysporum	OQ419360	OQ419285	OQ404959
FU-76	18988	Root	F. oxysporum	OQ419403	OQ419328	OQ405002
FU-82	18989	Stem	F. oxysporum	OQ419409	OQ419334	OQ405008
FU-13	18954	Tuber	F. proliferatum	OQ419355	OQ419280	OQ404954
FU-17	18955	Tuber	F. proliferatum	OQ419359	OQ419284	OQ404958
FU-62	18993	Tuber	F. proliferatum	OQ419395	OQ419320	OQ404994
FU-19	18958	Root	F. sambucinum	OQ419361	OQ419286	OQ404960
FU-16	–	Root	F. solani	OQ419358	OQ419283	OQ404957
FU-42	18996	Tuber	F. solani	OQ419384	OQ419309	OQ404983
FU-61	–	Root	F. solani	OQ419394	OQ419319	OQ404993
FU-93	–	Root	F. solani	OQ419421	OQ419346	OQ405020
FU-22	18959	Tuber	F. thapsinum	OQ419365	OQ419290	OQ404964
FU-55	19015	Tuber	Rectifusarium robinianum	OQ419390	OQ419315	OQ404989

^aDepartment of Plant Protection, Mycology Laboratory of the College of Agriculture and Natural Resources, University of Tehran, Iran.

^bAccession number of the strain in the Agro-Food Microbial Culture Collection of Institute of Sciences of Food Production, CNR, Italy (http://server.ispa.cnr.it/ITEM/Collection/).

Table 2.

Endophytic Alternaria strains used for molecular characterization, and accession numbers for their gene sequences in NCBI GeneBank database.

Origin and Strain No.(a)	Item No..(b)	Plant part	Section	Gene Bank Accession number
				Gpd	caM
Ardebil Province
AL-1	19017	Leaf	Alternaria	OQ419109	OQ419192
AL-2	19018	Stem	Alternaria	OQ419119	OQ419202
AL-3	19019	Stem	Alternaria	OQ419129	OQ419212
AL-4	–	Leaf	Alternaria	OQ419138	OQ419221
AL-11	–	Stem	Alternaria	OQ419111	OQ419194
AL-13	–	Stem	Alternaria	OQ419113	OQ419196
AL-15	–	Stem	Alternaria	OQ419114	OQ419197
AL-17	19025	Stem	Alternaria	OQ419116	OQ419199
AL-18	–	Stem	Alternaria	OQ419117	OQ419200
AL-21	–	Stem	Alternaria	OQ419120	OQ419203
AL-24	–	Leaf	Alternaria	OQ419123	OQ419206
AL-25	–	Stem	Alternaria	OQ419124	OQ419207
AL-26	19031	Stem	Alternaria	OQ419125	OQ419208
AL-27	–	Leaf	Alternaria	OQ419126	OQ419209
AL-28	–	Leaf	Alternaria	OQ419127	OQ419210
AL-30	19035	Stem	Alternaria	OQ419130	OQ419213
AL-33	–	Stem	Alternaria	OQ419132	OQ419215
AL-34	–	Stem	Alternaria	OQ419133	OQ419216
AL-37	19040	Leaf	Alternaria	OQ419136	OQ419219
AL-57	–	Leaf	Alternaria	OQ419157	OQ419240
AL-59	–	Stem	Alternaria	OQ419159	OQ419242
AL-67	–	Leaf	Alternaria	OQ419162	OQ419245
AL-70	–	Stem	Alternaria	OQ419165	OQ419248
AL-73	19033	Stem	Alternaria	OQ419168	OQ419251
AL-74	–	Stem	Alternaria	OQ419169	OQ419252
AL-80	19037	Stem	Alternaria	OQ419175	OQ419258
AL-85	–	Leaf	Alternaria	OQ419180	OQ419263
AL-95	19038	Stem	Alternaria	OQ419190	OQ419273
AL-96	19039	Leaf	Alternaria	OQ419191	OQ419274
AL-16	–	Leaf	Infectoria	OQ419115	OQ419198
AL-22	19042	Root	Infectoria	OQ419121	OQ419204
AL-23	–	Leaf	Infectoria	OQ419122	OQ419205
AL-51	–	Leaf	Infectoria	OQ419151	OQ419234
AL-12	19043	Stem	Ulocladioides	OQ419112	OQ419195
Kerman Province
AL-5	–	Leaf	Alternaria	OQ419149	OQ419232
AL-10	19022	Leaf	Alternaria	OQ419110	OQ419193
AL-19	19027	Stem	Alternaria	OQ419118	OQ419201
AL-29	19034	Stem	Alternaria	OQ419128	OQ419211
AL-32	19036	Leaf	Alternaria	OQ419131	OQ419214
AL-35	–	Leaf	Alternaria	OQ419134	OQ419217
AL-36	–	Stem	Alternaria	OQ419135	OQ419218
AL-39	–	Leaf	Alternaria	OQ419137	OQ419220
AL-40	–	Leaf	Alternaria	OQ419139	OQ419222
AL-41	–	Stem	Alternaria	OQ419140	OQ419223
AL-42	19017	Leaf	Alternaria	OQ419141	OQ419224
AL-43	19020	Leaf	Alternaria	OQ419142	OQ419225
AL-44	19021	Leaf	Alternaria	OQ419143	OQ419226
AL-45	–	Root	Alternaria	OQ419144	OQ419227
AL-46	19023	Stem	Alternaria	OQ419145	OQ419228
AL-47	–	Stem	Alternaria	OQ419146	OQ419229
AL-50	19024	Leaf	Alternaria	OQ419150	OQ419233
AL-52	19026	Leaf	Alternaria	OQ419152	OQ419235
AL-53	19028	Leaf	Alternaria	OQ419153	OQ419236
AL-55	19029	Stem	Alternaria	OQ419155	OQ419238
AL-56	19030	Leaf	Alternaria	OQ419156	OQ419239
AL-61	19032	Leaf	Alternaria	OQ419160	OQ419243
AL-63	–	Stem	Alternaria	OQ419161	OQ419244
AL-68	–	Leaf	Alternaria	OQ419163	OQ419246
AL-69	19044	Leaf	Alternaria	OQ419164	OQ419247
AL-71	-	Leaf	Alternaria	OQ419166	OQ419249
AL-72	19045	Root	Alternaria	OQ419167	OQ419250
AL-75	-	Leaf	Alternaria	OQ419170	OQ419253
AL-76	-	Leaf	Alternaria	OQ419171	OQ419254
AL-77	19046	Leaf	Alternaria	OQ419172	OQ419255
AL-78	19047	Leaf	Alternaria	OQ419173	OQ419256
AL-79	-	Root	Alternaria	OQ419174	OQ419257
AL-81	19048	Leaf	Alternaria	OQ419176	OQ419259
AL-82	19049	Stem	Alternaria	OQ419177	OQ419260
AL-83	-	Root	Alternaria	OQ419178	OQ419261
AL-84	-	Leaf	Alternaria	OQ419179	OQ419262
AL-86	-	Leaf	Alternaria	OQ419181	OQ419264
AL-87	19050	Leaf	Alternaria	OQ419182	OQ419265
AL-88	-	Leaf	Alternaria	OQ419183	OQ419266
AL-89	-	Leaf	Alternaria	OQ419184	OQ419267
AL-91	19051	Leaf	Alternaria	OQ419186	OQ419269
AL-92	19052	Leaf	Alternaria	OQ419187	OQ419270
AL-93	-	Leaf	Alternaria	OQ419188	OQ419271
AL-94	19053	Leaf	Alternaria	OQ419189	OQ419272
AL-48	19054	Stem	Ulocladioides	OQ419147	OQ419230
AL-49	19055	Stem	Ulocladioides	OQ419148	OQ419231
AL-54	19056	Leaf	Ulocladioides	OQ419154	OQ419237
AL-58	–	Stem	Ulocladioides	OQ419158	OQ419241
AL-90	19057	Stem	Ulocladioides	OQ419185	OQ419268

^aDepartment of Plant Protection, Mycology Laboratory of the College of Agriculture and Natural Resources, University of Tehran, Iran.

^bAccession number of the strain in the Agro-Food Microbial Culture Collection of Institute of Sciences of Food Production, CNR, Italy (http://server.ispa.cnr.it/ITEM/Collection/).

2.2. Morphological identification of endophytic Fusarium and Alternaria strains

Isolates of endophytes putatively assigned to Fusarium and Alternaria genera based on conidial morphology, were re-isolated from single spores and transferred on specific media to examine their macroscopic and microscopic features for the identification to the species level. Fusarium strains were incubated on PDA at 25 °C for 5–7 days with a 12/12 h day/night photoperiod to examine growth rate, colony features and production of chlamydospores; Carnation Leaf Agar (CLA) and Synthetic Nutrient Agar (SNA) were used for examination of macroconidia, chlamydospores, and phialides, according to Leslie and Summerell [24]. Morphological identification of Alternaria strains was carried out according to Simmons [25]. Small plugs of pure isolates were cultured on Potato Carrot Agar -PCA [26] and kept under 8/16 h of fluorescent light/dark cycle at 22 °C for 5–7 days. Then, colony features, sporulation patterns, conidial chains, shape, size and septa of conidia and primary and secondary conidiophores were examined.

2.3. Molecular characterization of Fusarium and Alternaria endophytic strains

Seventy-five (26 from Ardebil and 49 from Kerman) Fusarium strains (Table 1) and 83 Alternaria strains (34 from Ardebil and 49 from Kerman) (Table 2) were selected as representative of the Fusarium and Alternaria populations from different parts of potato plants, viz. leaves, stems, tubers, and roots. Selection of the strains was based on morphological features, sampling location (geographical area and farm), and potato plant tissue type.

For genomic DNA extraction and molecular analyses, the cryopreserved strains at ISPA-CNR were refreshed on PDA and then cultured on cellophane disks overlaid on PDA Petri dishes. After 3 days of growth at 25 °C, mycelia were scraped, transferred to 2 ml microtubes, frozen and lyophilized. Ten to fifteen mg of powdered lyophilized mycelium were used for DNA extraction by using the “Wizard Magnetic DNA Purification System for food” kit (Promega Corporation, Madison, WI), based on the manufacturer's protocol. The quantity and quality of extracted DNA were examined with Thermo-Scientific Nanodrop (LabX, Midland, ON, Canada), and by 0.8 % agarose gel electrophoresis, in comparison with a standard DNA (1 kb DNA Ladder, Fermentas GmbH).

2.3.1. Polymerase chain reaction (PCR) and sequencing

To molecularly identify Fusarium and Alternaria strains to the species level, and to evaluate the phylogenetic relationships within the two genera, a multi-locus sequencing approach was used. For the Fusarium strains, internal transcribed spacer regions (ITS), translation elongation factor 1α (tef1) and β-tubulin (tub2) genes were selected among the most informative genomic regions. For Alternaria strains, glyceraldephyde-3-phosphate dehydrogenase (gpd) and calmodulin (caM) gene fragments were chosen for the molecular analyses. The sequences of the primers used for PCR and the relevant references are summarised in Table 3.

Table 3.

Primers used for the molecular analysis of endophytic Fusarium spp. and Alternaria spp.

Gene	Primer	Sequence (5'-3')	Reference
ITS	ITS5	GGA AGT AAA AGT CGT AAC AAG G	[118]
	ITS4	TCC GCT TAT TGA TAT GC	[118]
tef1	EF1	ATG GGT AAG GAR GAC AAG AC	[119]
	EF2	GGA RGT ACC AGT SAT CAT GTT	[119]
tub2	Bt2a	GGT AAC CAA ATC GGT GCT TTC	[120]
	Bt2b	GGT AAC CAA ATC GGT GCT TTC	[120]
gpd	Gpd1	CAA CGG CTT CGG TCG CAT TG	[121]
	Gpd2	GCC AAG CAG TTG GTT GTG C-	[121]
CaM	CALDF1	AGC AAG TCT CCG AGT TCA AGG	[122]
	CALDR2	CTT CTG CAT CAY CTG GAC G	[122]

Each PCR reaction (total volume of 15 μl), containing 15 ng of genomic DNA, 300 nM each primer, 0.8 mM dNTPs, 1x PCR buffer and 0.6 U of Hot Start Taq DNA polymerase (Fisher Molecular Biology, Roma, Italy), was performed in the Mastercycler epgradient thermocycler (Eppendorf). The following PCR conditions were used: 95 °C for 2 min; 35 cycles for ITS, gpd, and caM, and 40 cycles for tef1 and tub2 of denaturation at 95 °C for 30 s, annealing at 52 °C for 40 s for ITS, 58 °C for 40 s for tef1, 58 °C for 30 s for tub2, gpd, and caM, extension step at 72 °C for 50 s, and final extension at 72 °C for 7 min. The quality and quantity of PCR products were examined and visualized by UV light after electrophoresis separation in 1 × TAE buffer, on 1.5 % agarose gel, in comparison with 100 bp DNA ladder (Invitrogen, Thermo Fisher Scientific). The PCR products were purified with enzymatic mixture of Exonuclease I/FastAP thermosensitive alkaline phosphatase (Thermo Fisher Scientific, Vilnius, Lithuania). Subsequent sequencing was performed with the Big Dye Terminator v3.1 Cycle Sequencing Ready Reaction Kit (Applied Biosystems, Foster City, CA, USA), according to the manufacturer's instructions. Each labelled product was purified by filtration through Sephadex G-50 (5%) (Sigma-Aldrich, Saint Louis, MO, USA) and sequenced in “ABI PRISM 3730 Genetic Analyzer” (Applied Biosystems, Foster City, CA, USA).

2.3.2. Phylogenetic analysis

The raw DNA sequences were edited, cleaned and assembled using the BioNumerics v. 5.1 software (Applied Maths, Kortrijk, Belgium). For each considered locus, partial FASTA sequences of endophytic Fusarium or Alternaria strains, and sequences of reference strains, were aligned using the ClustalW algorithm [27]. The sequences of the strains used as reference for Fusarium and Alternaria specieswere downloaded through the Fusarioid-ID database (http://www.fusarium.org) and the National Center for Biotechnology Information (NCBI).

For each fungal genus, the combined phylogenetic trees were built using the maximum likelihood method with the MEGA software version 7 [28]. The bootstrap analysis [29] was conducted to determine the confidence of internal nodes using a heuristic search with 1000 replicates, removing gaps. In the phylogenetic analyses of Fusarium, the strain Atractium crassum CBS 180.31 was used as outgroup taxon. Whereas, for the phylogenetic analyses of Alternaria, A. malorum CBS 135.31, belonging to the Chalastospora section, was utilized. Sequences obtained in this study were deposited in the GenBank database, with the accession numbers listed in Table 1, Table 2

2.4. Production of mycotoxins by the endophytic fusarium isolates

Twenty-six Fusarium isolates belonging to 6 species, viz. F. equiseti (2 isolates), F. nirenbergiae (5 isolates), F. oxysporum (6 isolates), F. nygamai (9), F. proliferatum (3 strains), and F. sambucinum (1 strain), mostly isolated from tubers, were tested for mycotoxin production in duplicate (Table 1). For production of mycotoxins, isolates were grown on 50 g of rice in 250 mL Erlenmeyer Flasks left to imbibe overnight with 30 mL (approx. 60% v/w) distilled water and then sterilized at 121 °C for 30 min. Each flask was inoculated with five pieces of a fresh fungal culture on PDA and incubated for 21 days at 25 °C, shaking every day. Then, the inoculated kernels were dried at 55 °C for 48 h and finely milled.

2.4.1. Extraction of mycotoxins from rice cultures

Based on the mycotoxigenicity of the Fusarium species considered, as reported by Munkvold [30], cultures were analyzed for the presence of beauvericin (BEA), enniatins (ENNs, namely ENN A, ENN A₁, ENN B, ENN B₁), fumonisins B (FBs, namely FB₁, FB₂ and FB₃), fusaric acid (FA), moniliformin (MON), the Type A trichothecenes diacetoxyscirpenol (DAS) and T-2 toxin (T-2), and the type B trichothecenes deoxynivalenol (DON), nivalenol (NIV) and their respective acetylates 3-AcDON, 15-AcDON and 4-AcNIV.

Two different extraction protocols were used. A multi-toxin extraction procedure was used for the extraction of BEA, ENNs, FBs, FA, and MON. One gram of ground rice culture was extracted with 5 mL of methanol/water (70:30, v/v) in an orbital shaker at 250 rpm for 60 min. The extract was filtered using Whatman No. 4 filters (Waters, Milford, MA, USA). One milliliter was diluted with 1 mL of water and then was filtered through 0.20 μm regenerated cellulose filter (Phenomenex, Torrance, CA, USA). The diluted culture extracts were used for the HPLC analyses.

The second procedure was used for the extraction of DON, as described in Quarta et al. [31]. One gram of ground rice culture was extracted in orbital shaker with 5 mL of acetonitrile/water (84:16, v/v) and 1% of acetic acid for 2 h. After filtration through filter paper (Whatman No. 4), 100 μL were diluted with 900 μL ultrapure water. The residue was filtered through 0.20 μm regenerated cellulose filter and analyzed by UHPLC/DAD.

2.4.2. Analyses of beauvericin and enniatins

Analyses of the cyclic hexadepsipeptides BEA and ENNs (ENN A, ENNA₁, ENN B, ENN B₁) were carried out by HPLC as described by Prosperini et al. [32] with minor modifications. One hundred microliters of the filtered extract of the culture (first extraction protocol) were injected into an HPLC apparatus (Agilent 1100 Series, Agilent Technology, Santa Clara, CA, USA) equipped with a binary solvent manager, a column heater set at 40 °C and a diode array (DAD) detector that was set at 205 nm wavelength. The analytical column was a Gemini (150 × 4.6 mm, 5 μm; Phenomenex) preceded by a SecurityGuard™ cartridge Gemini (4 × 3 mm, Phenomenex). The mobile phase was water as solvent A and acetonitrile as solvent B, eluted at a flow rate of 1 mL/min. A gradient elution was performed by changing the mobile phase composition. After 5 min at 70% eluent B, the proportion was set at 90% in 10 min, then kept constant for 3 min. The column was re-equilibrated with 70% eluent B for 5min. The retention times were: 11.3 min for BEA; 13.2 min for ENN A; 11.9 min for ENN A₁; 10.5 min for ENN B; 9.1 min for ENN B_1. The detection limits (LOD) based on signal-to-noise ratio of 3:1, were as follows: BEA = 0.01 μg/g, ENN A = 0.2 μg/g, ENN A₁ = 0.5 μg/g, ENN B = 0.06 μg/g, ENN B₁ = 0.07 μg/g. The mycotoxins were quantified by comparing peak areas with the calibration curves obtained with standard solutions.

2.4.3. Analysis of fumonisins

Analyses of FBs (FB₁, FB₂ and FB₃) were carried out by HPLC/FLD according to the procedure described by Haidukowski et al. [33] with minor modifications. The culture extract (first procedure) diluted and filtered (50 μL) was derivatized with o-phtaldialdehyde (50 μL) using the HPLC autosampler Agilent 1100 and injected after 3 min. The analytical column was a SymmetryShield RP18 (15 cm x 4,6 mm, 5 μm; Waters) set at 30 °C. The mobile phase consisted of a binary gradient applied as follows: the initial composition of the mobile phase 60% of (A) acetonitrile-water-acetic acid (B) acetonitrile-water-acetic acid (60/39/1, v/v/v), was kept constant for 5 min, then B solvent was linearly increased to 88% in 21 min, and kept constant for 4 min. The flow rate of the mobile phase was 1 mL/min. The fluorometric detector was set at wavelengths, ex = 335 nm, em = 440 nm. Retention time was about 16.4 min for FB₁, 25.4 min for FB₂ and 26.6 min for FB_3. Fumonisins were quantified by measuring peak areas and comparing them with the calibration curves obtained with standard solutions. LOD was 0.02 μg/g based on a signal-to-noise ratio of 3:1 for FB₁, FB₂ and FB₃.

2.4.4. Analysis of fusaric acid

FA was determined by HPLC using a 1100 Agilent instrument. The analytical column was a Gemini (150 × 4.6 mm, 5 μm; Phenomenex). The temperature of the column was maintained at 40 °C. Constant flow was set at 1.0 mL/min and mobile phase was 1 % formic acid as solvent A and methanol with 1% of formic acid as solvent B. The starting gradient ratio was 80% of solvent A and the final ratio was set at 30% of solvent A in 15 min, then kept constant for 2 min. FA was detected at 272 nm. Retention time was about 7.9 min, and the LOD was 0.25 μg/g.

2.4.5. Analysis of moniliformin

Analysis of MON was carried out according to the procedure described by Parich et al. [34] with minor modifications. Fifty microliters of the extract obtained with the first procedure were injected into the HPLC apparatus with a column thermostat set at 30 °C and a DAD detector set at 229 nm. The analytical column was a Symmetry C18 (150 × 4.6 mm, 5 μm; Waters). The mobile phase was a mixture of water/formic acid (99:1, v/v, solvent A) and methanol/formic acid (99:1, v/v, solvent B) eluted at a flow rate of 1 mL/min. A gradient elution was performed as follows: 50% B solvent for 5 min; then was linearly increased to 70% in 2 min, 50% at 12 min and kept constant for 5 min. The retention time of mycotoxin was about 3.9 min, and the LOD was 0.1 μg/g.

2.4.6. Analyses of trichothecenes type B (DON, NIV, 3-AcDON and 4-AcNIV)

Analysis of trichothecenes group B was carried out according to the procedure described by Pascale et al. [35]. Seven point 5 μL of extract were injected into a Waters Acquity UPLC/PDA system. The analytical column was Aquity UPLC BEH RP-C18 (2.1 × 100 mm, 1.7 μm; Waters) with an Acquity UPLC column in-line filter (0.2 μm). The column heater was set at 50 °C and the detector at 220 nm. The isocratic flow was set at 0.350 mL/min and the mobile phase was water/methanol (85:15 v/v). Under these analytical conditions the retention times of target toxins were NIV 1.3 min, DON 2.1 min, 4-AcNIV 4.3 min, 3-AcDON 9.9 min and 15-AcDON 10.5 min. The detection limits (LOD) were 0.02 μg/g for NIV, DON and 4-AcNIV, 0.05 μg/g for 3-AcDON and 15-AcDON. DON production was confirmed using DONTest™ immunoaffinity column (VICAM, Watertown, MA, USA) method. One gram of rice culture was extracted with 5 mL of water in an orbital shaker at 250 rpm for 60 min. The extract was filtered, and 2 mL was passed through immunoaffinity column. After washing with 5 mL of water, DON was eluted with 1.5 mL of methanol. Then, the extract was dried and solubilized with 250 μL of acetonitrile/water (10:90, v/v). Seven point five μL of extract was injected into to UPLC apparatus as described above. The detection limit (LOD) of DON based on signal-to-noise ratio of 3:1 was 0.02 μg/g. DON was quantified by comparing peak areas with a calibration curve obtained with standard solutions.

2.4.7. Analyses of trichothecenes type A (DAS and T-2 toxin)

Analysis of trichothecenes group A was carried out according to the procedure described by Pascale et al. [36] with minor modifications. Seven point 5 μL of extract (second protocol) were injected into a Waters Acquity UPLC system. The analytical column was Aquity UPLC BEH RP-C18 (2.1 × 100 mm, 1.7 μm; Waters) with an Acquity UPLC column in-line filter (0.2 μm). The column heater was set at 50 °C and the detector at 202 nm. The chromatographic separation was performed by a gradient elution water as solvent A and acetonitrile as solvent B. The initial composition of the mobile phase. (80 % solvent A, 20% solvent B) was kept constant for 2 min, then solvent B was linearly increased to 50% in 3 min, and kept constant for 1 min. The flow rate of the mobile phase was 0.7 mL/min. Retention times were 1.7 min for DAS and 4.2 min for T-2 toxin. LOD for DAS and T-2 toxin were 4 μg/kg and 8 μg/kg, respectively.

3. Results and discussion

3.1. Endophytic fusarium species

The anamorphic genus Fusarium Link is arguably the most agronomically important fungal genus associated to potato plants, since members of the genus are well-known for their phytopathogenicity and mycotoxin production. Out of almost 400 total fungal endophytic isolates from potato, 75 were belonging to the genus Fusarium, accounting for 18.75% of total isolates [10]. Fusaria were mostly isolated from the underground parts of potato plants (roots and tubers), suggesting some tissue specificity.

Seventy-five isolates were molecularly identified to species level based on the multi-locus sequencing of the genes tef1, tub2 and ITS, and assigned to 13 species, viz. F. brachygibosum, F. clavum, F. equiseti, F. flocciferum, F. incarnatum, F. nirenbergiae, F. nygamai, F. oxysporum, F. proliferatum, F. redolens, F. sambucinum, F. solani and F. thapsinum (Table 1). Five Fusarium species, namely F. oxysporum, F. redolens, F. semitectum, F. solani, and F. sporotrichioides were previously reported as endophytic in potato [13]. In our survey, we did not find any endophytic strains of either F. semitectum or F. sporotrichioides.

The phylogenetic analysis of the concatenated sequences of 1337 sites resulted in a phylogenetic combining dataset comprising 92 taxa, including 75 Fusarium field strains, 16 Fusarium reference sequences and the strain Atractium crassum CBS 180.31 as outgroup taxon. Four out of the 79 endophytic Fusarium strains isolated (FU-57, -28, −56, −80) were not included in phylogenetic analyses since they did not give PCR products of tef1. However, the comparison of tub and ITS sequences through the Blast N program (http://www.ncbi.nlm.nih.gov/) allowed us to assign these strains to the genus Plectospharella. A high homology (more than 99%) was found with the deposited sequences of P. cucumerina (syn. Fusarium tabacinum) strains.

The phylogenetic tree, rooted to outgroup taxon, was resolved in 10 well-separated clades, supported by high significance (bootstrap values more than 96), as reported in Fig. 1. Twenty-two out of 75 strains and F. oxysporum CBS 144,134 and F. nirembergiae CBS 129.24 reference strains grouped together in a well-supported clade corresponding to the Fusarium oxysporum species complex (FOSC, clade A). In this clade, a very low genetic variability was observed. Indeed, all the strains showed high level of similarity among them and with the two reference strains included in the analyses. In particular, the strains ITEM-18945, -18979, −18980, −18974 showed 100% homology with the F. nirembergiae reference strain. Clade B, referred to as Fusarium redolens species complex (FRSC), grouped the reference strain F. redolens NRRL 25600 together with the strains ITEM-18990, -18991, and −18992 with high level of homology. Twenty-four endophytic Fusarium strains, identified as belonging to Fusarium fujikuroi species complex (FFSC), were grouped in clades C and D. Four Fusarium strains, showing 100% of homology among them, clustered with F. proliferatum reference strain (clade C). In clade D, two clusters were distinguished: the first cluster contained a single strain (ITEM-18959) that showed high similarity with F. thapsinum CBS 776.96 and the second cluster contained majority of the strains (19 out of 20) along with F. nygamai CBS 749.97 reference strain.Only two strains (ITEM-18957 and -18961) clustered with F. brachygibbosum NH-1 reference strains (clade E) and one single strain with F. sambucinum CBS 146.95 strain (Clade F). Eleven endophytic Fusarium strains included in the clade G, shared high genetic diversity. This clade included strains assigned to different species of Fusarium incarnatum-equiseti species complex (FIESC). One single strain (ITEM-18963) was genetically closely related to the reference strain F. incarnatum FIESC 23. Three strains (ITEM-18966, -18968, and −18967) grouped with the reference strains F. equiseti FIESC 14. Finally, in a well-supported group, 7 Fusarium strains clustered with the reference strain F. equiseti FIESC 5. Likewise, large diversity was observed in the clade I, in which 10 endophytic F. solani strains clustered with the two F. solani strains used as references of the Fusarium solani species complex (FSSC). On the other hand, two strains (ITEM-18960 and -19015) showed a high similarity with F. flocciferum CBS 821.68, member of Fusarium tricinctum species complex, and F. robinianum (syn. Rectifusarium robinianum), respectively (clades H and J).

Fig. 1.

Phylogenetic tree generated by Maximum Likelihood method (bootstrap 1000 replicates) from combined DNA sequences of tub, tef1, and ITS fragments of 75 endophytic Fusarium strains isolated from different organs of potato plants, in Iran. Atractium crassum strain CBS 180.31 was used as an outgroup. Isolates with an asterisk were used for mycotoxin analysis.

Ninety percent of endophytic Fusaria strains were isolated from roots and tubers. Only four species, viz. F. nygamai, F. oxysporum and F. solani, were isolated from stems and only one isolate of F. proliferatum was obtained from potato leaves (Fig. 2). At a whole, F. nygamai, member of FFSC, F. oxysporum and F. nirembergiae, both members of FOSC, F. solani and F. equiseti (FIESC), were the most abundant species (Fig. 3a). In particular, F. nygamai was predominant in south Kerman (Fig. 3b), and F. equiseti, F. oxysporum and F. solani were more frequently isolated in Ardebil (Fig. 3c). Therefore, F. nygamai seems to be more adapted to dry and hot climate conditions, while F. oxysporum and F. solani can apparently colonize potato plants under diverse climate conditions.

Fig. 2.

Occurrence and distribution of endophytic Fusarium species in different organs of potato plants.

Fig. 3.

Frequency of endophytic Fusarium species in potato plants from Iran (a), detailed for South Kerman (b) and Ardebil (c) regions. The species complexes are reported in brackets: FFSC: Fusarium fujikuroi species complex, FOSC: F. oxysporum species complex, FIESC: F. incarnatum-equiseti species complex.

The species F. brachygibbosum, F. clavum, F. flocciferum and F. nirenbergiae are herein reported as endophytic for the first time. So far, F. brachygibbosum has been reported as a plant pathogen of different crops, including wheat [37], maize [38], date palm [39,40], Cannabis sativa [41], 2019) and several medicinal plants [42,43], but not potato. Fusarium clavum was previously isolated from crop plants cultivated under both conventional and organic farming [44]. The species causes post-harvest contamination of wheat, barely, and maize [45] and associated to leaf spot disease in vegetable plants [46] and leaf wilt in date palm [47]. Fusarium flocciferum is a common species in temperate regions, soil, roots, fruits, stems and twigs of various plants [48]. Fusarium nirenbergiae is a plant pathogenic species and an agent of vascular diseases such as wilting of common bean [49], passion fruit (Passiflora edulis Sims) [50] and ornamental plants [51], but also reported as the agent of saffron corm rot disease [52].

The species F. equiseti, F. incarnatum, F. proliferatum, F. nygamai, F. sambucinum and F. thapsinum were reported to be endophytes in different plants, but not in potato. Particularly, F. equiseti was isolated from leaf of the medicinal plant Sophora tonkinensis [53], barley roots [54], and Salicornia bigelovii [55]. Fusarium incarnatum is the causal agent of postharvest fruit rot in muskmelon (Cucumis melo) [56], and of a fruit disease of bell peppers [57], but the species is reportedly endophytic in the mangrove plant Aegiceras corniculatum [58]. Fusarium proliferatum is a widespread phytopathogen in a number of major crops, including rice, wheat, maize, garlic, asparagus, date palm, and Chinese chive [59], and an endophyte in wheat [60]. On the other hand, few studies report the endophytic associations of F. nygamai with medicinal plants, namely Alhagi graecorum, Cressa cretica, Citrullus colocynthis, Tamarix nilotica, Achillea fragrantissima, Artemisia sieberi, and Neurospora retusa [42], and rice root [61]. Fusarium sambucinum, which is recognized as a major agent of dry rot of potato tubers [24] has also been reported as an endophyte from Nicotiana tabacum [62](Zhang et al., 2019). Fusarium thapsinum causes stalk rot and grain mold of sorghum [24], but it was also found to be endophytic in the same plant [63].

3.2. Endophytic Alternaria species

The phylogenetic combining dataset comprised 82 taxa, including 83 Alternaria field strains, and 9 additional Alternaria reference sequences, among which A. malorum CBS 135.31 which was used as outgroup taxon. The Alternaria isolates were molecularly identified at the species level based on the multi-locus sequencing of the genes gpd and CaM (Table 3).

The phylogenetic tree, rooted to outgroup taxon, was resolved in 3 well-separated clades, supported by high bootstrap values (Fig. 4). Most of the strains (73 out of 83) were included in the clade A, referred as Alternaria section. In this clade, high genetic homology was observed among the strains. Indeed, except for two strains (ITEM-19051 and -19023), 54 Alternaria strains clustered with A. alternata EGS34-016 and A. tenuissima EGS34-015 reference strains; 13 strains were highly similar to A. lini CBS 106.34 reference strain, and four strains clustered, in a well-separated group (bootstrap value 84) with A. arborescens 39–128 reference strain. In clade B, referred as Ulocladioides section, 6 strains grouped with two members of this section, A. consortialis and A. utra. The strains ITEM-9043, AL-58, and ITEM-19057 were very close to A. utra ATCC18040 reference strain; whereas ITEM-19054, -19055 and −19057, showing 100% similarity among them, were very close to A. consortialis CBS 201.67 reference strain. Finally, four strains (ITEM-19042, AL-51, -16, and −23) grouped with A. infectoria EGS27-193 and A. viburni EGS 49–147 reference strains in the Infectoriae section (clade C).

Fig. 4.

Phylogenetic tree generated by Maximum Likelihood method (bootstrap 1000 replicates) from combined DNA sequences of gpd and caM fragments of 83 endophytic Alternaria strains isolated from different organs of potato plants, in Iran. Alternaria malorum strain CBS 135.31 was used as an the outgroup.

Alternaria spp. are ubiquitous fungi that include endophytic, saprophytic, and pathogenic species commonly found in soil, air, food commodities, and on decaying plant tissues [64]. The pathogenic species of Alternaria cause major losses on a wide range of crops such as cereals, oil crops, ornamentals, vegetables and fruits [65]. In our survey, out of almost 400 total fungal endophytic isolates, 96 were identified as Alternaria spp. [10], that is 24.06 % of total isolates. Ninety-five percent of endophytic Alternaria strains were isolated from the above-ground plant parts (stems and leaves). Only five isolates were originated from roots, four of which belonged to the Alternaria section and one to the Infectoriae section (Fig. 4). It appears that the different climatic conditions in the two Iranian regions of Ardebil and Kerman do not affect significantly the distribution of endophytic Alternaria; in both the regions the Alternaria section was predominant, supposedly due to greater tolerance to diverse climate conditions.

The species A. alternata, A. tenuissima, and A. lini were the most abundantly occurring species in this study. However, in the last decade Alternaria taxonomy has been deeply revised, based on genetic and genomic investigations. Sequencing analyses of the most informative genomic regions led to the conclusion that species that were morphologically different could not be distinguished genetically [66,67]. Therefore, around 35 morpho-species, including A. alternata, A. tenuissima and A. lini, should be considered one single species and synonymized as A. alternata [66]. Recently, this finding has been supported also in a genome-wide study based on phylogenomic analyses of Alternaria species belonging to Alternaria Section [68]. Anyway, morpho-species names continue to be reported in several studies. Previously, only three Alternaria species, namely A. alternata, A. solani, and A. brassicola were reported as endophytic in potato [11,13]. Alternaria alternata is a cosmopolitan species with a wide host range worldwide, including cereals [69,70], vegetables [64,71] and halophyte plants [72]. It is pathogenic in many important crops and also lives in asymptomatic symbiosis as an endophyte of many plants [73], including potato, Nicotiana spp. [74], forest tree species [75], and Juncus spp [76]. This is the first report of A. lini as an endophytic species. Alternaria tenuissima has been widely reported as endophyte from different plants [77,78], but it is herein reported as an endophyte of potato for the first time. Alternaria arborescens was isolated as endophyte from root tissue of Combretum latifolium [79], and is also reported as a phytopathogen of potato and tomato [64,71,80,81], and a human pathogen [82]. Alternaria atra (previously known as Ulocladium atra [65] was reported as an endophyte of common yew (Taxus baccata L.) in Iran [83]. Alternaria consortialis (previuosly knwon as U. consortilis [65], was reported as endophyte of several plants in Iran, including Prunus trees [84], and spinach [85], and as a phytopathogen of declined Persian oak trees in Iran [86] and on date palm showing leaf spot in Tunisia [87]. Alternaria viburni, previously known as Lewia viburni [65], is reported as endophyte from timothy (Phleum pratense) and perennial ryegrass [88]. This species belongs to the Infectoriae Section, which is phylogenetically divergent from the Alternaria Section and characterized by a great genetic variability within and among its species members [67].

While Alternaria spp. are mainly regarded as plant pathogenic fungi that cause diseases of aerial parts of plants and are also capable of producing potentially harmful metabolites [89], beneficial effects of endophytic Alternaria on plant growth have also been reported [90]. Whether the endophytic Alternaria strains associated with potato plants may turn to be pathogenic and under which conditions this may occur, or conversely may have beneficial effect on potato plants, remains to be clarified and is worth of further investigation.

3.3. Mycotoxin production by potato endophytic fusarium strains

The toxigenicity of twenty-six Fusarium strains from potato root and tubers was assessed. The strains tested belonged to 6 species, viz. F. equiseti (2 strains), F. nirenbergiae (5 strains), F. oxysporum (6 strains), F. nygamai (9 strains), F. proliferatum (3 strains), and F. sambucinum (1 strain). Twenty-one out of 26 strainswere able to produce one or more mycotoxin (Table 4).

Table 4.

Production of mycotoxins by representative isolates of major endophytic Fusarium species from potato root and tuber. None of the isolates tested produced either the type B trichothecenes deoxynivalenol, nivalenol or their respective acetylates or the type A trichothecenes diascetoxyscirpenol and T-2 toxin.

Species	Strain	Mycotoxin production (μg/g)a
		Enniatins				BEA	Fumonisins			MON	FA
		B	B1	A	A1		B1	B2	B3
F. nirenbergiae	ITEM 18972	n.d.	n.d.	n.d.	n.d.	396	n.d.	n.d.	n.d.	82	n.d.
	ITEM 18973	n.d.	n.d.	n.d.	n.d.	101	n.d.	n.d.	n.d.	63	n.d.
	ITEM 18983	n.d.	n.d.	n.d.	n.d.	1382	n.d.	n.d.	n.d.	66	n.d.
	ITEM 18981	n.d.	n.d.	n.d.	n.d.	952	n.d.	n.d.	n.d.	31	n.d.
	ITEM 18982	n.d.	n.d.	n.d.	n.d.	332	n.d.	n.d.	n.d.	27	n.d.
F. oxysporum	ITEM 18988	n.d.	n.d.	n.d.	n.d.	46	n.d.	n.d.	n.d.	21	n.d.
	ITEM 18941	n.d.	n.d.	n.d.	n.d.	n.d.	n.d.	n.d.	n.d.	105	n.d.
	ITEM 18946	n.d.	n.d.	n.d.	n.d.	n.d.	n.d.	n.d.	n.d.	n.d.	n.d.
	ITEM 18987	n.d.	n.d.	n.d.	n.d.	525	n.d.	n.d.	n.d.	n.d.	n.d.
	ITEM 18942	n.d.	n.d.	n.d.	648	n.d.	n.d.	n.d.	n.d.	n.d.	156
	ITEM 18952	n.d.	n.d.	n.d.	446	7	n.d.	n.d.	n.d.	n.d.	n.d.
F. proliferatum	ITEM 18954	n.d.	n.d.	n.d.	n.d.	n.d.	n.d.	n.d.	n.d.	n.d.	n.d.
	ITEM 18955	n.d.	n.d.	n.d.	n.d.	235	62	23	11	n.d.	n.d.
	ITEM 18994	n.d.	n.d.	n.d.	n.d.	58	3	2	2	n.d.	n.d.
F. nygamai	ITEM 18952	n.d.	n.d.	n.d.	n.d.	n.d.	5	2	2	n.d.	n.d.
	ITEM 18947	n.d.	n.d.	n.d.	n.d.	n.d.	n.d.	n.d.	n.d.	n.d.	n.d.
	ITEM 18948	n.d.	n.d.	n.d.	n.d.	n.d.	n.d.	n.d.	n.d.	26	n.d.
	ITEM 18951	n.d.	n.d.	n.d.	n.d.	n.d.	105	29	24	67	n.d.
	ITEM 19012	n.d.	n.d.	n.d.	n.d.	4190	7946	4693	4333	538	n.d.
	ITEM 19013	n.d.	n.d.	n.d.	n.d.	41	49	18	13	4	n.d.
	ITEM 19014	n.d.	n.d.	n.d.	n.d.	n.d.	126	5	4	26	n.d.
	ITEM 18949	n.d.	n.d.	n.d.	n.d.	458	n.d.	n.d.	n.d.	75	n.d.
	ITEM 18950	n.d.	n.d.	n.d.	n.d.	n.d.	n.d.	n.d.	n.d.	4	n.d.
F. equiseti	ITEM 18966	n.d.	n.d.	n.d.	n.d.	n.d.	n.d.	n.d.	n.d.	n.d.	n.d.
	ITEM 18967	n.d.	n.d.	n.d.	n.d.	n.d.	n.d.	n.d.	n.d.	n.d.	n.d.
F. sambucinum	ITEM 18958	n.d.	n.d.	n.d.	n.d.	21	n.d.	n.d.	n.d.	n.d.	n.d.

^aamounts are the mean of two replicated cultures on sterile rice; BEA = beauvericin; MON = moniliformin; FA = fusaric acid; n.d. = not detected.

None of the tested strains produced either type A or type B trichothecenes. Two out of two F. equiseti strains, 1/6 F. oxysporum, 1/3 F. proliferatum, and 1/9 F. nygamai did not produce any of the other tested toxins. All the 5 strains of F. nirenbergiae produced both BEA (from moderate to high quantity, 101.0–1382.5 μg/g) and MON (2.9–82.1 μg/g). Fusarium oxysporum showed a more variable mycotoxin profile: ENN A1 was produced by 2/6 strains, BEA by 3/6, MON by 2/6 and FA by 1/6. BEA was found in combination with MON in a single strain; ENN A1 was found in combination with either FA (1 strain) or BEA (1 strain). Among the F. proliferatum strains, 1/3 did not produce any of the mycotoxins analyzed, while the other 2 strains produced both BEA and FBs (B1, B2, and B3). In F. nygamai, which was the most numerous group, 3/9 strains produced BEA, 4/9 produced FBs (B1, B2, and B3), and 7/9 produced MON. All the fumonisins-producing strains also produced MON and 2 of them produced BEA, as well. One isolate produced BEA in combination with MON. The most toxigenic strain was ITEM-19012, which produced the highest quantities of FBs (7946.5, 4693.0 and 4333.0 μg/g of B1, B2, and B3 respectively), along with the highest quantities of both BEA (4190.0 μg/g) and MON (538.4 μg/g). The only toxin produced by F. sambucinum was BEA.

Amongst the regulated mycotoxins, only FBs were produced by potato endophytic strains of Fusarium spp., while trichothecenes were not found in cultures of two representative isolates of F. equiseti and one F. sambucinum isolate, reportedly the only two trichothecene-producing species among the ones analyzed [30]. Fumonisins were produced by strains of F. proliferatum and F. nygamai, in all of the three most important forms B1, B2, and B3. Particularly, the ITEM 19012 F. nygamai strainproduced significantly more FBs than the other strains. This particular strain also produced the highest levels of both BEA and MON. Indeed, members of F. fujikuroi complex are known to produce multiple mycotoxins, including BEA, ENNs, FUMs, fusaproliferin, FA, fusarins, and MON [18]. Co-occurrence of MON with the cyclodepsipeptides BEA and ENNs has been also reported in F. sporotrichioides, F. equiseti, F. avenaceum and F. oxysporum [18].

Besides the regulated mycotoxins, we also investigated production of some “emerging” mycotoxins, which are neither routinely determined, nor legislatively regulated because of their lesser occurrence or uncertain toxicity in vivo. ENNs and BEA are toxic compounds similar for both structure and mode of action. They are cyclic hexadepsipeptides consisting of three d-α-hydroxy-isovaleryl-(2-hydroxy-3-methylbutanoic acid) alternating with three amino acid units and are biosynthesized by the non-ribosomal multifunctional enzymes enniatin and beauvericin synthetases. They both have ionophoric properties; they can form stable lipophilic complexes with cations and transport them into cell membrane and form cation-selective channels in membranes, thus impairing the membrane functions [91,92]. In potato, ENNs are virulence factors within the plant-pathogen interaction [93]. They have phytotoxic effects that include wilting and necrosis of tissues and are produced in vivo in potato tuber tissue infected by Fusariun spp [23]. However, their function as virulence factor seems to be host dependent [94]. Likewise, BEA showed strong phytotoxicity to tomato protoplasts, conceivably by inducing release of ascorbate from the cytosol to the apoplast of plant cell, with a consequent increase in cell wall plasticity that facilitates pathogen penetration [95]. BEA, along with FA, were contaminants of banana plants and fruits infected by F. oxysporum f. sp. cubense and proved to have phytotoxic effects on banana protoplasts [96]. In that study, virulence of F. oxysporum f. sp. cubense isolates correlated well with toxin accumulation. BEA was produced, in combination with unusual forms of ENNs, by fifteen F. oxysporum and two not-identified Fusarium sp. strains, out of twenty-eight Fusarium isolates from potato samples in Korea [97]. ENNs and BEA exhibit different toxic effects in vitro, including cytotoxicity and necrotic and pro-apoptotic effects to animal cell lines, but toxicity in vivo is generally low. This is thought to be due to their rapid metabolization rather than low bioavailability. MON is a phytotoxin, first isolated by Cole et al. [98]. Despite its name, nowadays MON is thought to be produced by a few Fusariun species, including F. oxysporum, F. proliferatum, F. nygamai, F. equiseti and F. sambucinum, but not by F. moniliforme [30]. MON is extremely soluble in water, due to its polarity, thus it is easily translocated through the plant. MON causes growth inhibition, necrosis, and chlorosis in many plants [99]. MON has a selective cytotoxicity in vitro, but shows severe effects in vivo, with symptoms of acute intoxication in test animals that include muscular weakness, respiratory stress, myocardial degeneration, and histopathological changes in kidneys, lungs, and pancreas and ultimately coma and death. FA has long been known as a wilt toxin [100] and is regarded as a virulence factor in plant tracheo-fusariosis caused by F. oxysporum ff. spp. in different crops [[101], [102], [103]], including potato [104]. FA is moderately toxic to animals, but it may potentiate the effect of other Fusarium toxins.

Although the above toxins are mainly regarded as phytotoxins that participate in the pathogenicity process, because of their occurrence in vivo and toxicity in animal models, they are emerging as possible concern for food safety. A recent study showed that BEA was the emergent mycotoxin with the highest prevalence in feed and feed ingredients, followed by ENNs [105]. Naturally, MON often co-occurs with ENNs and BEA, but also with trichothecenes, since several trichothecene-producing species produce MON, including F. acuminatum, F. culmorum, F. equiseti, and F. sporotrichioides [106]. FA is widespread on corn and corn-based food and feeds, frequently in association to other mycotoxigenic Fusarium species [107]. These toxins may contribute to the overall health risk of contaminated foods or feeds, either because of their direct toxic effects or because of additive or synergistic interaction with other co-occurring mycotoxins.

All the isolates we analyzed for mycotoxin production were endophytes of potato plants, recovered from healthy tubers or roots. Endophytic vs. pathogenic behavior of fungi is the outcome of a finely tuned balance, which depends on a molecular cross-talking between plant and symbiont [108]. Even if the molecular and physiological aspect of this interaction and its determinants are not completely clear, it is widely accepted that endophytism is a continuum which results from an equilibrium between fungal virulence and plant defense, and that endophytic behavior might revert to pathogenic, if one or more factors change [109]. In this sense, the chance that endophytic Fusaria become pathogenic and produce mycotoxins in tubers cannot be ruled out. Logrieco et al. [110] examined the production of zearalenone, zearalenols, trichothecenes and MON in cultures of some Fusarium species isolated from rotted potato tubers. They found that strains able to produce mycotoxins on autoclaved rice and maize kernels were not capable to produce toxins on fresh potato under any of the conditions tested. Conversely, Ellner [22] reported that potato tubers artificially infected with F. sambucinum were contaminated with the trichothecene toxin diacetoxyscirpenol in concentrations up to 200 μg/tuber, depending on the susceptibility of the cultivar tested. The toxin was also detected in tubers with no apparent disease symptoms. Likewise, Delgado et al. [21]studied the accumulation and diffusion of trichothecenes in potato tubers affected with dry rot caused by F. graminearum during storage. They found accumulation of either DON or NIV, depending on the F. graminearum genotype, in rotten tissue but not in the surrounding tissue. These findings suggest that contamination of potato tubers with mycotoxins in the field or at post-harvest, due to a change in lifestyle of endophytic microflora, is a potential risk that should be carefully considered and that is worth of further studies, along with possible means of control.

4. Conclusions

Plant-microbe endophytic association may bring a vast range of beneficial effects to the host plants. Endophytes produce biologically active metabolites with diverse activities, which interact with their host's physiology and ecology in different ways. These compounds can benefit host plants by protecting them against insect herbivores [111,112], by inhibiting plant pathogen growth and by eliciting host immune system against pathogens attacks [113]. On the other hand, some endophytes may actually become detrimental to their hosts, especially under stress conditions [109], or they may become pathogenic to crop plants other than their original host [114]. Thus, some endophytic fungi can be actually regarded as latent pathogens [8]. The chance that under certain conditions the endophytes of potato may change their lifestyle and become pathogenic to potato or other crop plants, should be considered.

Members of Fusarium are found worldwide, predominantly as soil inhabitants and plant pathogens, with the existence of nonpathogenic isolates [115,116]. In addition, some members of Fusarium, including some herein found to live endophytically in potato, are known to produce numerous toxins, some of which are regarded as virulence factors [93] or emerging mycotoxins (viz. ENNs, BEA, MON and FA) and others are regulated mycotoxins (viz. fumonisins), which may pose safety concerns for consumers. It is not yet well-established if, in certain conditions, the endophytic Fusaria may switch to a parasitic lifestyle and, in that case, if accumulation of toxins in the edible tubers can occur in the field or during storage. The findings of this and previous work [10], also open the prospect to conceivable strategies aimed at controlling pathogens of potato and, particularly, for biological control of mycotoxigenic Fusaria. Classical biological control is founded on the concept that most effective biocontrol agents should be sought amongst the coevolved antagonists of the target pathogen, which are supposed to have the highest specificity and biocontrol activity. In this regard, the report of the occurrence of fungi belonging to Trichoderma and Clonostachys, two well-known genera of biocontrol agents [117], as endophytes of potato in Iran [10] paves the way for studies on their use for sustainable control of potentially pathogenic endophytic fungi and for prevention of mycotoxin contamination of potatoes.

Funding

This work was supported by the Iran National Science Foundation (INSF) [grant number 98015699].

Additional information

No additional information is available for this paper.

Data availability statement

Data will be made available on request.

The data and supportive information are available within the article. Data associated with the study have been deposited into publicly available repositories. Sequences were deposited in the GenBank database; strains were deposited in the microbial ITEM collection (http://server.ispa.cnr.it/ITEM/Collection/).

CRediT authorship contribution statement

Nasim Alijani Mamaghani: Writing – original draft, Visualization, Resources, Investigation, Formal analysis, Data curation. Mario Masiello: Writing – original draft, Visualization, Validation, Methodology, Investigation, Formal analysis, Data curation. Stefania Somma: Writing – review & editing, Visualization, Validation, Methodology, Investigation, Formal analysis, Data curation. Antonio Moretti: Writing – review & editing, Supervision, Resources, Methodology, Conceptualization. Hossein Saremi: Writing – original draft, Supervision, Resources, Project administration, Funding acquisition, Conceptualization. Miriam Haidukowski: Writing – review & editing, Validation, Resources, Methodology, Investigation, Formal analysis, Data curation. Claudio Altomare: Writing – review & editing, Writing – original draft, Visualization, Supervision, Funding acquisition, Conceptualization.

Declaration of competing interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Miriam Haidukowski reports travel was provided by Iran National Science Foundation. Miriam Haidukowski reports financial support, administrative support, article publishing charges, and equipment, drugs, or supplies were provided by Institute of Sciences of Food Production, National Research Council, Italy. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.