Taxonomy of fungal complex causing red-skin root of Panax ginseng in China.

BACKGROUND: Red-skin root of Asian ginseng (Panax ginseng) significantly reduces the quality and limits the production of ginseng in China. The disease has long been thought to be a noninfectious physiological disease, except one report that proved it was an infectious disease. However, the causal agents have not been successfully determined. In the present study, we were to reveal the pathogens that cause red-skin disease.
METHODS: Ginseng roots with red-skin root symptoms were collected from commercial fields in Northeast China. Fungi were isolated from the lesion and identified based on morphological characters along with multilocus sequence analyses on internal transcription spacer, β-tubulin (tub2), histone H3 (his3), and translation elongation factor 1α (tef-1α). Pathogens were confirmed by inoculating the isolates in ginseng roots.
RESULTS: A total of 230 isolates were obtained from 209 disease samples. These isolates were classified into 12 species, including Dactylonectria sp., D. hordeicola, Fusarium acuminatum, F. avenaceum, F. solani, F. torulosum, Ilyonectria mors-panacis, I. robusta, Rhexocercosporidium panacis, and three novel species I. changbaiensis, I. communis, and I. qitaiheensis. Among them, I. communis, I. robusta, and F. solani had the highest isolation frequencies, being 36.1%, 20.9%, and 23.9%, respectively. All these species isolated were pathogenic to ginseng roots and caused red-skin root disease under appropriate condition.
CONCLUSION: Fungal complex is the causal agent of red-skin root in P. ginseng.

Introduction

Asian ginseng (Panax ginseng) is a perennial herb, mainly cultivated for pharmaceutical purpose in China and Korea [1], [2]. Dry roots of ginseng have been used for more than 4000 years to stimulate metabolism, hence maintaining and improving health of human beings [1], [2]. The value of roots is determined by their size, shape, and overall appearance [3]. Ginseng cultivation requires multiple years, and generally four- to six-year-old roots are harvested for sale. In such a long time of cultivation in the field, roots are vulnerable to many soilborne diseases [4]. Red-skin root is the most common and serious problem in Northeast China, which is a major ginseng production area [5], [6], [7]. Red-skin root can occur in all ages of ginseng, but disease severity is more in later growing years, particularly after the fourth year [8]. Disease incidence can be up to 80% in heavily occurring fields. Red-skin root symptom greatly reduces root marketability by up to 40% [9].

Red-skin root is usually characterized by less fibrous roots and reddish-brown to orangish-brown discolored lesions with irregular shapes and margins at the crown of the tap root or areas forming lateral roots, sometimes even whole roots in the fields with heavy diseases. Typically, the superficial lesion can be easily scraped off, resulting in the exposure of inner white healthy tissue. Since red-skin root was first described in the 1960s in China [10], most researchers have treated it as a noninfectious physiological disease due to lack or excess of mineral nutrition and soil pH or moisture; they distinguished it from rusty root diseases [3], [10], [11], [12], [13]. However, Shang et al [7] reported that healthy ginseng roots can be infected by red-skin roots in the field. Meanwhile, the abiotic factors, including soil humidity and temperature, and fertilizers were not determinants but only accelerate the disease development [7]. Unfortunately, the causal agents have not been determined by the authors. The limited knowledge of red-skin root hinders the development of effective management strategies.

Rusty root of American ginseng has symptoms similar to red-skin root and has been well documented [14]. Rusty root is characterized by small or quite large reddish-brown areas at the crown of the tap root that can be easily scraped off that exposes the inner white healthy tissue [14]. Rusty root is caused by weak pathogens including Cylindrocarpon destructans/Ilyonectria radicicola species complex [14], [15]. In China, pathogens of ginseng Cylindrocarpon root rot, rusty root rot, or rust rot diseases are divided into highly virulent species such as C. destructans and C. panacis and less virulent species such as C. panacicola and C. obtusisporum [9], [16], [17], [18]. Furthermore, the taxon of C. destructans species complex has been classified into 12 novel species by morphological and multigene analysis [19]. In addition to Cylindrocarpon species, Fusarium species and Rhexocercosporidium panacis have been reported to be the causal agents of American ginseng rusty root [20], [21], [22], [23].

Our preliminary data led us to speculate that Asian ginseng red-skin root disease was an infectious disease caused by weak pathogens. To prove this hypothesis, we were to 1) isolate potential pathogenic microorganisms from ginseng grown in Northeast China, 2) identify the pathogen complex using multilocus analysis and morphological characteristics, and 3) confirm the pathogenesis of the isolates.

Materials and methods

Isolates

Two hundred and nine fresh ginseng roots with red-skin root symptoms (Fig. 1) were collected from 13 commercial fields in 9 counties of Northeast China between June 2012 and September 2013. Ginseng roots were washed in running tap water and blotted to dry. Small pieces of red-skin tissue were surface disinfested with 0.62% NaClO for 3 min, rinsed with sterile distilled water, air-dried, and cut into about 5-mm (in length) pieces. The tissue was placed on potato dextrose agar (PDA) amended with 100 μg/ml of chloramphenicol and 100 μg/ml of tetracycline [24]. Plates were incubated at 25°C for up to 2 weeks. Single spores or single hyphal tips were transferred to PDA plates for later use. All isolates were stored at −80°C. Representative isolates were deposited in China General Microbiological Culture Collection Center (CGMCC, link: http://www.cgmcc.net/), Beijing, China.

Fig. 1

Symptoms of red-skin disease of Panax ginseng. (A and B) On 6-year-old roots. (C and D) On 3-year-old roots. (E) Cross-section of a 3-year-old root (F) Comparison with a 6-year-old healthy ginseng root.

Morphological observation

Fungal isolates were grown at 22°C on PDA and oatmeal agar in the dark for 2 weeks before observation. Culture characteristics, including texture, density, color, growth front, transparency, and zonation, were visually examined [25]. Colony colors observed from the surface and reverse, both top and back, were described using the color chart of Rayner [26].

Microscopic observation of morphology of fungal isolates was conducted using cultures grown on PDA and synthetic nutrient agar [27] under continuous n-UV light (400–315 nm). A Nikon Eclipse (D v4.50, Nikon, Tokyo) 80i light microscope equipped with a Digital Sight DS-L2 camera (Nikon, Tokyo) and NIS-Element software were used to capture digital images. For each isolate, at least 30 measurements were obtained for each structure. Measurements are given as minimum (lower limit of a 95% confidence interval), average, and maximum (upper limit of a 95% confidence interval). Based on morphology observation, Fusarium isolates were identified into genus level.

DNA extraction, polymerase chain reaction amplification, DNA sequencing, and multigene phylogenies

For each isolate, total genomic DNA was isolated from mycelium harvested from the 7-day-old colony grown on PDA at 25°C, using the FastDNA Plant Kit (Biomed Co. Ltd, Beijing, China) and the Precellys 24 Technology homogenizer (Bertin Technology, France) according to the manufacturer's instructions.

Partial gene sequences were obtained by using the following protocols. Primers pair ITS1 and ITS4 were used for partial internal transcription spacer (ITS) [28], CYLH3F and CYLH3R for partial his3 [29], EF1 and EF2 for partial tef-1α [30], and BT3 (CCCCTGATTCTACCCCGC) and BT4 (CTGACCGAAGACGAAGTTGTC) for partial tub2 designed in this study. Sequences of polymerase chain reaction amplicons were assembled and edited with Chromas 1.5 (Technelysium Pty Ltd, Queensland, Australia) and DNAMAN 6.0 (Lynnon BioSoft, Quebec, Canada). Newly obtained sequences were deposited in GenBank (Table 1). Sequence alignments were generated using MAFFT, version 7 (Katoh & Standley 2013, Japan). For Fusarium isolates, only partial sequences of the tef-1α gene were amplified and blasted on the GenBank database for identification.

Table 1

Cylindrocarpon-like isolates used in the phylogenetic analyses.

Species	Isolate no.1)	Substrate	Locality	Collector	GenBank accession no.2)
					ITS	tub2	his3	tef1-α
Campylocarpon fasciculare	CBS 112613	Vitis vinifera	South Africa	F. Halleen	AY677301	AY677221	JF735502	JF735691
C. pseudofasciculare	CBS 112679	Vitis vinifera	South Africa	F. Halleen	AY677306	AY677214	JF735503	JF735692
Cylindrodendrum album	CBS 301.83	Fucus distichus	Canada	R. C. Summerbell	KM231764	KM232021	KM231484	KM231889
C. album	CBS 110655	Soil	The Netherlands	F.X. Prenafeta-Boldú	KM231765	KM232022	KM231485	KM231890
C. alicantinum	CBS 139518	Eriobotrya japonica	Spain	J. Armengol	KP456014	KP400578	KP639555	KP452501
C. alicantinum	Cyl-8	Eriobotrya japonica	Spain	J. Armengol	KP456015	KP400579	KP639556	KP452502
C. hubeiensis	CBS 124071	Rhododendron sp.	China	W. P. Wu, W. Y. Zhuang, Y. Nong	FJ560439	FJ860056	KR909093	HM054090
C. hubeiensis	CBS 129.97	Viscum album	France	W. Gams	KM231766	KM232023	KM231486	KM231891
Dactylonectria alcacerensis	CBS129087	Vitis vinifera	Portugal	C. Rego, H. Oliveira	JF735333	AM419111	JF735630	JF735819
D. alcacerensis	Cy134	Vitis vinifera	Spain	J. Armengol	JF735332	AM419104	JF735629	JF735818
D. anthruriicola	CBS 564.95	Anthurium sp.	The Netherlands	R. Pieters	JF735302	JF735430	JF735579	JF735768
D. estremocensis	CBS 129085	Vitis vinifera	Portugal	C. Rego, T. Nascimento	JF735320	JF735448	JF735617	JF735806
D. estremocensis	CPC 13539	Picea glauca	Canada	R. C. Hamelin	JF735330	JF735458	JF735627	JF735816
D. hordeicola	CBS 162.89	Hordeum vulgare	The Netherlands	M. Barth	AM419060	AM419084	JF735610	JF735799
D. hordeicola	3S07	Panax ginseng	China	X. H. Lu	MF350482	MF350428	MF350455	MF350509
D. macrodidyma	CBS 112615	Vitis vinifera	South Africa	F. Halleen	AY677290	AY677233	JF735647	JF735836
D. macrodidyma	CBS 112601	Vitis vinifera	South Africa	F. Halleen	AY677284	AY677229	JF735644	JF735833
D. novozelandica	CBS 113552	Vitis sp.	New Zealand	R. Bonfiglioli	JF735334	AY677237	JF735633	JF735822
D. novozelandica	CBS 112608	Vitis vinifera	South Africa	F. Halleen	AY677288	AY677235	JF735632	JF735821
D. pauciseptata	CBS 120171	Vitis sp.	Slovenia	M. Žerjav	EF607089	EF607066	JF735587	JF735776
D. pauciseptata	CBS 100819	Erica melanthera	New Zealand	H. M. Dance	EF607090	EF607067	JF735582	JF735771
D. pinicola	CBS 173.37	Pinus laricio	Germany	H. W. Wollenweber	JF735319	JF735447	JF735614	JF735803
D. pinicola	CBS 159.34	Pinus laricio	UK: England	T. R. Peace	JF735318	JF735446	JF735613	JF735802
D. torresensis	CBS 129086	Vitis vinifera	Portugal	A. Cabral	JF735362	JF735492	JF735681	JF735870
D. torresensis	CBS 119.41	Fragaria sp.	The Netherlands	H. C. Koning	JF735349	JF735478	JF735657	JF735846
D. vitis	CBS 129082	Vitis vinifera	Portugal	C. Rego	JF735303	JF735431	JF735580	JF735769
Dactylonectria sp.	CGMCC 3.18786 = J711	Panax ginseng	China	X. H. Lu	MF350479	MF350425	MF350452	MF350506
Dactylonectria sp.	YJ212	Panax quinquefolius	China	X. H. Lu	MF350480	MF350426	MF350453	MF350507
Dactylonectria sp.	YJ515	Panax quinquefolius	China	X. H. Lu	MF350481	MF350427	MF350454	MF350508
Ilyonectria capensis	CBS 132815	Protea sp.	South Africa	C. M. Bezuidenhout	JX231151	JX231103	JX231135	JX231119
I. capensis	CBS 132816	Protea sp.	South Africa	C. M. Bezuidenhout	JX231160	JX231112	JX231144	JX231128
I. changbaiensis	CGMCC 3.18789 = 4404	Panax ginseng	China	X. H. Lu	MF350464	MF350410	MF350437	MF350491
I. changbaiensis	72R2	Panax ginseng	China	X. H. Lu	MF350465	MF350411	MF350438	MF350492
I. changbaiensis	11R8	Panax ginseng	China	X. H. Lu	MF350466	MF350412	MF350439	MF350493
I. changbaiensis	1506	Panax ginseng	China	X. H. Lu	MF350467	MF350413	MF350440	MF350494
I. changbaiensis	1803	Panax ginseng	China	X. H. Lu	MF350468	MF350414	MF350441	MF350495
I. changbaiensis	306	Panax ginseng	China	X. H. Lu	MF350469	MF350415	MF350442	MF350496
I. changbaiensis	320	Panax ginseng	China	X. H. Lu	MF350470	MF350416	MF350443	MF350497
I. changbaiensis	3S10	Panax ginseng	China	X. H. Lu	MF350471	MF350417	MF350444	MF350498
I. communis	CGMCC 3.18788 = 1512	Panax ginseng	China	X. H. Lu	MF350456	MF350402	MF350429	MF350483
I. communis	J410	Panax ginseng	China	X. H. Lu	MF350457	MF350403	MF350430	MF350484
I. communis	71R2	Panax ginseng	China	X. H. Lu	MF350458	MF350404	MF350431	MF350485
I. communis	J101	Panax ginseng	China	X. H. Lu	MF350459	MF350405	MF350432	MF350486
I. communis	301	Panax ginseng	China	X. H. Lu	MF350460	MF350406	MF350433	MF350487
I. communis	H207	Panax ginseng	China	X. H. Lu	MF350461	MF350407	MF350434	MF350488
I. communis	J710	Panax ginseng	China	X. H. Lu	MF350462	MF350408	MF350435	MF350489
I. communis	J305	Panax ginseng	China	X. H. Lu	MF350463	MF350409	MF350436	MF350490
I. coprosmae	CBS 119606	Metrosideros sp.	Canada	G. J. Samuels	JF735260	JF735373	JF735505	JF735694
I. crassa	CBS 139.30	Lilium sp.	The Netherlands	W. F. van Hell	JF735275	JF735393	JF735534	JF735723
I. crassa	CBS 158.31	Narcissus sp.	The Netherlands	W. F. van Hell	JF735276	JF735394	JF735535	JF735724
I. cyclaminicola	CBS 302.93	Cyclamen sp.	The Netherlands	M. Hooftman	JF735304	JF735432	JF735581	JF735770
I. destructans	CBS 264.65	Cyclamen persicum	Sweden	L. Nilsson	AY677273	AY677256	JF735506	JF735695
I. europaea	CBS 129078	Vitis vinifera	Portugal	C. Rego	JF735294	JF735421	JF735567	JF735756
I. europaea	CBS 537.92	Aesculus hippocastanum	Belgium	V. Demoulin	EF607079	EF607064	JF735568	JF735757
I. gamsii	CBS 940.97	Soil	The Netherlands	J. T. Poll	AM419065	AM419089	JF735577	JF735766
I. leucospermi	CBS 132809	Leucospermum sp.	South Africa	Zhuang, Y. Nong	JX231161	JX231113	JX231145	JX231129
I. leucospermi	CBS 132810	Protea sp.	South Africa	C. M. Bezuidenhout	JX231162	JX231114	JX231146	JX231130
I. liliigena	CBS 189.49	Lilium regale	The Netherlands	M. A. A. Schippers	JF735297	JF735425	JF735573	JF735762
I. liliigena	CBS 732.74	Lilium sp.	The Netherlands	G. J. Bollen	JF735298	JF735426	JF735574	JF735763
I. liriodendri	CBS 110.81	Liriodendron tulipifera	USA	J.D. MacDonald, E.E. Butler	DQ178163	DQ178170	JF735507	JF735696
I. liriodendri	CBS 117526	Vitis vinifera	Portugal	C. Rego	DQ178164	DQ178171	JF735508	JF735697
I. lusitanica	CBS 129080	Vitis vinifera	Portugal	N. Cruz	JF735296	JF735423	JF735570	JF735759
I. mors-panacis	CBS 306.35	Panax quinquefolium	Canada	A. A. Hildebrand	JF735288	JF735414	JF735557	JF735746
I. mors-panacis	CBS 124662	Panax ginseng	Japan	Y. Myazawa	JF735290	JF735416	JF735559	JF735748
I. mors-panacis	11R9	Panax ginseng	China	X. H. Lu	MF350477	MF350423	MF350450	MF350504
I. palmarum	CBS 135754	Howea forsteriana	Italy	G. Polozzi	HF937431	HF922608	HF922620	HF922614
I. palmarum	CBS 135753	Howea forsteriana	Italy	G. Polozzi	HF937432	HF922609	HF922621	HF922615
I. panacis	CBS 129079	Panax quinquefolium	Canada	K. F. Chang	AY295316	JF735424	JF735572	JF735761
I. protearum	CBS 132811	Protea sp.	South Africa	C. M. Bezuidenhout	JX231157	JX231109	JX231141	JX231125
I. protearum	CBS 132812	Protea sp.	South Africa	C. M. Bezuidenhout	JX231165	JX231117	JX231149	JX231133
I. pseudodestructans	CBS 129081	Vitis vinifera	Portugal	C. Rego	AJ875330	AM419091	JF735563	JF735752
I. pseudodestructans	CBS 117824	Quercus sp.	Austria	E. Halmschlager	JF735292	JF735419	JF735562	JF735751
I. qitaiheensis	CGMCC 3.18787 = H309	Panax ginseng	China	X. H. Lu	MF350472	MF350418	MF350445	MF350499
I. qitaiheensis	J919	Panax ginseng	China	X. H. Lu	MF350473	MF350419	MF350446	MF350500
I. robusta	CBS 308.35	Panax quinquefolium	Canada	A. A. Hildebrand	JF735264	JF735377	JF735518	JF735707
I. robusta	CBS 129084	Vitis vinifera	Portugal	N. Cruz	JF735273	JF735391	JF735532	JF735721
I. robusta	J906	Panax ginseng	China	X. H. Lu	KM015300	KM015297	KM015299	KM015298
I. rufa	CBS 153.37	Sand dune	France	F. Moreau	AY677271	AY677251	JF735540	JF735729
I. rufa	CBS 640.77	Abies alba	France	F. Gourbière	JF735277	JF735399	JF735542	JF735731
I. strelitziae	CBS 142253	Strelitzia reginae	Italy	D. Aiello	KY304649	KY304755	KY304621	KY304727
I. strelitziae	CBS 142254	S. reginae	Italy	D. Aiello	KY304651	KY304757	KY304623	KY304729
I. venezuelensis	CBS 102032	Bark	Venezuela	A. Y. Rossman	AM419059	AY677255	JF735571	JF735760
I. vredenhoekensis	CBS 132807	Protea sp.	South Africa	C. M. Bezuidenhout	JX231155	JX231107	JX231139	JX231123
I. vredenhoekensis	CBS 132808	Protea sp.	South Africa	C. M. Bezuidenhout	JX231159	JX231111	JX231143	JX231127

Epi-type and ex-type isolates indicated in bold. Sequences generated in this study indicated in italics

CBS: CBS-KNAW Fungal Biodiversity Centre, Utrecht, The Netherlands; CGMCC: China General Microbiological Culture Collection Center, Beijing, China.

ITS: the internal transcribed spacer region and intervening 5.8S nrRNA; tub2: β-tubulin;his3: histone H3; tef1-α: translation elongation factor 1-alpha.

The most suitable substitution model was determined based on jModelTest [31]. Maximum likelihood (ML) analyses including 500 bootstrap replicates were run using RAxML BlackBox web server (Gamma model of rate heterogeneity) [32]. Bayesian analyses were performed using MrBayes, version 3.1.2 [33]. A Markov chain Monte Carlo algorithm of four chains was initiated in parallel from a random tree topology with a heating parameter set at 0.2. The Markov chain Monte Carlo analyses lasted until the average standard deviation of split frequencies were below 0.01. The sample frequency was set to 100, and the first 25% of trees were removed as burn-in. Campylocarpon fasciculare and C. pseudofascicul were designated as the outgroup for all analyses. The resulting trees were obtained using FigTree, version 1.4.2, (Andrew Rambaut, UK) and annotated using Adobe Illustrator CS5.

Pathogenicity

Pathogenicity test was carried out on detached ginseng roots in vitro and also roots growing in potting soil inoculated with randomly selected isolates from each species. For test in vitro, fresh 3-year-old roots were dug from fields and gently washed with tap water, and roots with blemishes were discarded. Healthy roots were surface sterilized as described previously and placed on moist filter paper in an enamel tray. Mycelial plugs (5 mm in a diameter) cut from the margin of actively growing colonies were placed on ginseng roots with the mycelial side facing down to roots that had either a premade hole or not, about 2 to 4 plugs per root, and four replicated roots were inoculated for each isolate with noncolonized agar plugs as control. The tray was sealed with plastic film to prevent desiccation and incubated in the dark at 20 ± 1°C. After 10 days of inoculation, pathogens were isolated from every root with symptomatic lesions and mock-inoculated control roots as described previously to confirm the inoculated isolates. For test in greenhouse, healthy, fresh, 2-year-old roots were obtained as described previously and planted in pots (2.5 L) with sterilized soil. Three ginseng plants were kept in each pot. Conidia suspensions were made by flooding actively sporulating cultures on PDA plates with sterile distilled water and filtering with sterilized lens-wiping paper to remove mycelia. Conidia concentrations were measured and adjusted to 1 × 105 conidia/mL using a hemocytometer. Then, 10 μL of the suspension was drenched to one pot, and four pots were inoculated for each isolate. Sterile distilled water was used to drench control plants. The pots were maintained in greenhouse under 75% shade cloth. After 85 days, all roots were dug out and gently washed with tap water. Then, disease symptoms were observed, and pathogens were reisolated from roots with symptomatic lesions and also mock-inoculated control roots to confirm pathogen isolates.

Results

Isolation and identification

In total, 230 fungal isolates were obtained from ginseng roots with typical red-skin root symptoms (Fig. 1 and Table 2). In most cases, one species was isolated per lesion, but there were 21 isolations from which more than one species were obtained from a single lesion. Based on colony morphology and conidial characteristics, 74 isolates were preliminarily identified as Fusarium species, 151 isolates were Cylindrocarpon-like species (Fig. 2, Fig. 3, Fig. 4, Fig. 5). The other 5 isolates had been described as Rhexocercosporidium panacis previously [34]. For Fusarium isolates, 4 were classified as F. acuminatum, 7 were F. avenaceum, 55 were F. solani, and 8 were F. torulosum, based on partial DNA sequences of tef1-α. For Cylindrocarpon-like isolates, 7 species were identified, including Dactylonectria hordeicola, Dactylonectria sp., I. mors-panacis, I. robusta, I. changbaiensis, I. communis and I. qitaiheensis.

Table 2

Fungal isolates recovered from Panax ginseng with red-skin disease symptoms in Northeastern China.

Location (county, city, province)	Number of isolation	Dactylonectria sp.	D. hordeicola	Fusarium acuminatum	F. avenaceum	F. solani	F. torulosum	Ilyonectria changbaiensis	I. communis	I. mors-panacis	I. qitaiheensis	I. robusta	Rhexocercosporidium panacis	Total
Tonghe, Harbin, Heilongjiang	9					7						2	2	11
Bei'an, Heihe, Heilongjiang	7			1		3						6		10
Qiezihe, Qitaihe, Heilongjiang	9					4					1	3		8
Fusong, Baishan, Jilin	20					12			8			3		23
Changbai, Baishan, Jilin	114		1	3	7	18	8	15	61	1		3		117
Jiaohe, Jilin, Jilin	5					2			3			2	2	9
Antu, Yanbian, Jilin	20	1				3			11			8		23
Hunchun, Yanbian, Jilin	6											6		6
Ji'an, Tonghua, Jilin	19					6					1	15	1	23
Total	209	1	1	4	7	55	8	15	83	1	2	48	5	230

Fig. 2

Phylogenetic tree of Cylindrocarpon-like isolates based on the analysis of combined 4 genes. Branches with BS = 100% and PP = 1.00 are thickened and in red. Braches with BS ≥ 80% and PP ≥ 0.95 are thickened and in green. The phylogram is rooted with Campylocarpon fasciculare (CBS 112613) and C. pseudofasciculare (CBS 112679).

Fig. 3

Morphological characters of Ilyonectria changbaiensis (CGMCC 3.18789). (A–C) Macroconidia and microconidia. (D and E) Conidiophores. (F) Chlamydospores. Bar = 10 μm.

Fig. 4

Morphological characters of Ilyonectria communis (CGMCC 3.18788). (A–C) Microconidia and macroconidia. (D and E) Chlamydospores. (F and H) Conidiophores. Bar = 10 μm.

Fig. 5

Morphological characters of Ilyonectria qitaiheensis (CGMCC 3.18787). (A–C) Macroconidia and microconidia. (D and E) Conidiophores. (F and G) Chlamydospores. Bar = 10 μm.

Phylogenetic analysis of Cylindrocarpon-like isolates

Polymerase chain reaction amplicons of approximately 450 bases for tub2 and his3, 500 bases for ITS, and 800 bases for tef1-α were obtained for 22 isolates sequenced. The combined alignment of the ITS, tub2, his3 and tef1-α had a total length of 1894 characters including alignment gaps (520 for ITS, 454 for tub2, 449 for his3, and 471 for tef1-α). An analysis by jModelTest proposed the best model TIM2+I+G. ML analysis resulted in a single best ML tree with likelihood = −13331.071129 by using RAxML. Bayesian analysis lasted 330000 generations, and the consensus tree was calculated from 4689 trees left after 250 trees were discarded as burn-in.

The phylogenetic tree based on the combined analysis of four loci (Fig. 2) classified the 82 taxa into 39 species, fulfilling the requirements of genealogical concordance phylogenetic species recognition [35]. All the Cylindrocarpon-like isolates obtained from P. ginseng were grouped into seven highly supported clades (with maximum likelihood bootstrap (ML-BS) of 100% and bayesian inference posterior probabilities (BI-PP) 1.0). Three of the clades, I. robusta, I. mors-panacis, and D. hordeicola, have been described previously. The other four clades represent three novel Ilyonectria species, including I. communis, I. changbaiensis, and I. qitaiheensis, and one novel Dactylonectria species.

Phylogenetic analyses were also conducted on the individual locus and yielded trees with similar topology, but with rearrangement in the order of some clades. Of all loci used, ITS is the least informative region. The trees of both his3 and tub2 could separate all the species, but some clades had lower supporting values than those of the combined tree. Tree of tef1-α could resolve all species except I. communis and I. robusta, which were divided into two separate groups. The alignments and phylogenetic trees were deposited in TreeBASE (S23012).

Taxonomy

The morphological characteristics well supported by phylogenetic analyses revealed that isolates 3S07, 11R9, and J906 were D. hordeicola, I. mors-panacis, and I. robusta, respectively. Based on the phylogenetic and morphological data, three novel taxa in the genera Ilyonectria are named in this study, and one new species in Dactylonectria will be treated separately.

X. Lu & W. Gao, sp. nov

MycoBank MB823893.

(Fig. 3)

Etymology: Named after the county of Changbai, Jilin Province, China, where the isolates were collected.

Diagnosis: Ilyonectria changbaiensis can be distinguished from the phylogenetically closely related I. communis, I. crassa, I. panacis, I. pseudodestructans, and I. rufa in shorter and thicker 3-septate macroconidia.

Type: China: Jilin Province, Baishan, Changbai, on roots of Panax ginseng, Oct 2012, X. Lu (CGMCC 3.18789 = 4404 - holotype).

Description: Conidiophores simple or complex. Simple conidiophores arising laterally or terminally from aerial mycelium, solitary, dichotomously branched or unbranched or commonly branched with up to three phialides, 0- to 3-septate, 46- to 72-μm long, phialides monophialidic, cylindrical, tapering toward the apex, 16- to 62-μm long, 2.5- to 3.5-μm wide at base, 5 μm at the widest point, 1.5–2.5 μm near the aperture. Complex conidiophores aggregated in small sporodochia, repeatedly and irregularly branched, phialides more or less cylindrical, tapering toward the apex, 16- to 33-μm long, 2 to 3-μm wide at the base, 1.5–2.5 μm wide at the apex. Macroconidia formed on both types of conidiophores, 1- to 3-septate, straight, cylindrical with both ends more or less broadly rounded, mostly without a visible hilum; 1-septate, (16.0-)22.8-23.4-23.9(-33.0) × (4.0-)6.2-6.3-6.5(-8.0) μm, with a length:width ratio of 2.4–5.2; 2-septate, (22.0-)27.7-28.3-28.9(-36.0) × (5.0-)6.6-6.8-6.9(-8.0) μm, with a length:width ratio of 3.1–5.0; 3-septate, (25.0-)30.0-30.7-31.5(-38.0) × (6.0-)6.7-6.9-7.0(-8.0)μm, with a length:width ratio of 3.3–5.4. Microconidia 0- to 1-septate, more or less straight, with a laterally displaced hilum; aseptate microconidia globose to subglobose, (4.0-)7.4-7.7-8.1(-12.0) × (3.0-)3.8-3.9-4.0(-5.0) μm, with a length:width ratio of 1.3–3.3; one-septate microconidia ellipsoidal to ovoid, (9.0-)11.7-12.0-12.4(-16.0) × (3.0-)4.1-4.2-4.3(-5.0) μm, with a length:width ratio of 2.0–4.0. Chlamydospores globose to subglobose to ellipsoidal, 7-16 × 7-14 μm, smooth but often appearing rough due to deposits, thick-walled, terminal or intercalary, in chains or in clumps, hyaline, becoming medium brown, and formed abundantly in mature colonies. Sexual state not observed.

Culture characteristics: Mycelium felty with strong density. Surface on PDA was golden red, zonation was absent, and reverse was dark brown to yellow brown. Colony diameter was 51–61 mm at 22°C after 7 days. Hardly grew at 4°C and 30°C (no more than 3 mm colony diameter after 7 days).

Additional culture examined: China, Jilin Province, Baishan, Changbai, on roots of Panax ginseng, Oct 2012, X. Lu (320 &72R2).

X. Lu & W. Gao, sp. nov.

MycoBank MB823894.

(Fig. 4)

Etymology: “communis” = Latin for “common”. The name is given because this is the commonest Ilyonectria species causing Panax ginseng red-skin root disease in Northeast China.

Diagnosis: Ilyonectria communis can be distinguished from the phylogenetically closely related I. crassa, I. pseudodestructans, I. rufa, and I. panacis, with the former having more phialides of a simple conidiophore and thicker 3-septate macroconidia.

Type: China: Jilin Province, Baishan, Changbai, on roots of Panax ginseng, Oct 2012, X. Lu (CGMCC 3.18788 = 1512 - holotype).

Description: Conidiophores simple or complex. Simple conidiophores arising laterally or terminally from aerial mycelium, solitary, unbranched or frequently branched with up to four phialides, 0- to 3-septate, 58- to 94-μm long, phialides monophialidic, cylindrical, tapering toward the apex, 18- to 32-μm long, 2.1- to 3.3-μm wide at base, 5 μm at the widest point, 1.4–2.3 μm near the aperture. Complex conidiophores aggregated in small sporodochia, repeatedly and irregularly branched, phialides more or less cylindrical, tapering toward the apex, 16- to 33-μm long, 2–3 μm wide at the base, 1.5- to 2.5-μm wide at the apex. Macroconidia formed on both types of conidiophores, 1- to 3-septate, straight and frequently minutely curved, cylindrical or sometimes typically minutely widening toward the tip, mostly with a visible hilum; 1-septate, (13.0-)23.3-23.9-24.3(-34.0) × (4.0-)6.2-6.3-6.4(-9.0) μm, with a length:width ratio of 3.3–4.2; 2-septate, (20.0-)28.9-29.4-29.8(-38.0) × (5.0-)6.4-6.5-6.6(-9.0)μm, with a length:width ratio of 4.0–5.0; 3-septate, (23.0-)29.8-30.3-30.8(-42.0) × (5.0-)6.8-6.9-7.0(-9.0) μm, with a length:width ratio of 4.0–5.0. Microconidia 0- to 1-septate, ellipsoidal to ovoid to subcylindrical, more or less straight, without a visible hilum; aseptate microconidia, (5.0-)8.7-8.9-9.1(-13.0) × (3.0-)4.0-4.1-4.2(-6.0) μm, with a length:width ratio of 1.7–2.5; one-septate microconidia, (6.0-)12.3-12.6-12.8(-18.0) × (3.0-)4.5-4.6-4.7(-7.0)μm, with a length:width ratio of 2.3–3.2. Chlamydospores globose to subglobose to ellipsoidal, 6–25 × 6–15 μm, smooth but often appearing rough due to deposits, thick-walled, terminal or intercalary, in chains or in clumps, and also in the cells of the macroconidia, becoming medium brown, and formed abundantly in mature colonies. Sexual state not observed.

Culture characteristics: Mycelium felty with average density. Surface on PDA was gray yellow, and that on reverse was dark gray brown to light golden. Colony diameter was 46–56 mm at 22°C after 7 days. Hardly grew at 4°C and 30°C (no more than 3 mm colony diameter after 7 days).

Additional culture examined: China, Jilin Province, Baishan, Changbai, on roots of Panax ginseng, Oct 2012, X. Lu (71R2, H207, J101, 314-2 & J710).

Notes: Ilyonectria communis differs from the phylogenetically closely related I. crassa, I. pseudodestructans, I. rufa, and I. panacis with respect to the number of phialides of a simple conidiophore and the diameter of 3-septate macroconidia [19]. Two or three phialides of a simple conidiophore are common for I. communis, but conidiophores are unbranched or sparsely branched, up to two phialides for I. crassa, I. pseudodestructans, I. rufa, and I. panacis [19]. The average thickness of the 3-septate macroconidia of I. communis (av. = 30.3 × 6.9 μm) was more than the average thickness of those of I. crassa (av. = 35.1 × 5.7 μm), I. pseudodestructans (av. = 35.2 × 6.0 μm), I. rufa (av. = 29.9 × 5.7 μm), and I. panacis (av. = 33.1 × 5.6 μm) [36].

X. Lu & W. Gao, sp. nov.

MycoBank MB823895

(Fig. 5)

Etymology: Named after the city of Qitaihe, Heilongjiang Province, China, where it was collected.

Diagnosis: Ilyonectria qitaiheensis can be distinguished from the phylogenetically closely related I. liliigena and I. gamsii in macroconidia mostly minutely curved with the tip end.

Type: China: Heilongjiang Province, Qitaihe, Qiezihe, on roots of Panax ginseng, Oct 2013, X. Lu (CGMCC 3.18787 = H309 - holotype).

Description: Conidiophores simple or complex. Simple conidiophores arising laterally or terminally from aerial mycelium, solitary, unbranched or sparsely branched with up to two phialides, 0- to 3-septate, 46- to 132-μm long, phialides monophialidic, cylindrical, tapering toward the apex, 15- to 40-μm long, 1.8- to 3.0-μm wide at base, 4.0 μm at the widest point, 1.2–2.2 μm near the aperture. Complex conidiophores aggregated in small sporodochia, repeatedly and irregularly branched, phialides more or less cylindrical, tapering toward the apex. Macroconidia formed on both types of conidiophores, 1- to 3-septate, straight or mostly minutely curved with the tip end, cylindrical or sometime typically minutely widening toward the tip, mostly with a visible hilum; 1-septate, (15.0-)21.8-22.8-23.9(-34.0) × (4.0-)5.1-5.3-5.5(-7.0)μm, with a length:width ratio of 3.6–4.9; 2-septate, (21.0-)27.9-28.9-29.9(-37.0) × (4.0-)5.6-5.8-6.0(-8.0)μm, with a length:width ratio of 4.3–5.7; 3-septate, (22.0-)29.3-30.7-32.0(-44.0) × (5.0-)5.7-5.9-6.1(-8.0) μm, with a length:width ratio of 4.4–5.8. Microconidia 0- to 1-septate, globose to ellipsoidal to subcylindrical, more or less straight, mostly with a visible hilum; aseptate microconidia, (3.0-)7.9-8.4-8.8(-12.0) × (3.0-)3.4-3.6-3.8(-6.0)μm, with a length:width ratio of 1.0–3.7; one-septate microconidia, (9.0-)10.5-11.1-11.6(-14.0) × (3.0-)3.7-3.9-4.2(-6.0) μm, with a length:width ratio of 2.5–3.3. Chlamydospores globose to subglobose to ellipsoidal, sparely, 8–14 × 7–20 μm, smooth but often appearing rough due to deposits, thick-walled, terminal or intercalary, in chains or in clumps, becoming medium brown, and formed abundantly in mature colonies. Sexual state not observed.

Culture characteristics: Mycelium felty with average density and sparse mycelium. Surface on PDA was gray yellow, and that on reverse was gray brown to dark golden. Colony diameter was 52–60 mm at 22°C after 7 days. Hardly grew at 4°C and 30°C (no more than 2 mm colony diameter after 7 days).

Additional culture examined: China, Jilin Province, Baishan, Changbai, on roots of Panax ginseng, Oct 2012, X. Lu (J919).

Notes: Ilyonectria qitaiheensis differs from the phylogenetically closely related I. liliigena and I. gamsii with respect to macroconidia mostly minutely curved with the tip end [19].

For test in vitro, all the isolates tested in Ilyonectria, Dactylonectria, and Fusarium were pathogenic to ginseng roots (Fig. 6). For most isolates inoculated on punctured roots, rot lesions were restricted around the point of inoculation without expansion, and around rot lesions, red-skin root symptoms showed. For most isolates on nonpunctured roots, only red-skin root symptoms were observed and the disease lesions were superficial and solid. For the isolates in F. avenaceum (Fig. 6I) and F. torulosum (Fig. 6K), soft rot symptoms expanded clearly and deep into the cortex. For test in whole plant, all the isolates tested were pathogenic to cause red-skin roots (Fig. 7). Roots infected by I. mors-panacis showed larger disease lesions and less lateral roots than roots infected by other pathogens (Fig. 7E). Besides red-skin root symptoms, root infected by F. acuminatum showed dry rot lesion on taproots (Fig. 7H). All isolates were recovered from symptomatic roots and confirmed by analyzing DNA sequence of histone H3 gene separately. The mock-inoculated control roots remained symptomless, and no Dactylonectria, Ilyonectria, or Fusarium isolates were isolated. The inoculation experiments were repeated, and both trials showed the same results. Besides Cylindrocarpon-like species and Fusarium species, we have found that R. panacis is also a causal agent of red-skin root of ginseng in our previous report [34]. Among these species, I. communis (Fig. 6D), I. robusta (Fig. 6G), and F. solani (Fig. 6J) were the commonest species with isolation frequency of 36.1%, 20.9%, and 23.9%, respectively.

Fig. 6

Symptoms of red-skin root disease induced by in vitro inoculation on detached Panax ginseng roots with the fungi. (A) Dactylonectria sp. (B) D. hordeicola. (C) Ilyonectria changbaiensis. (D) I. communis. (E) I. mors-panacis. (F) I. qitaiheensis. (G) I. robusta. (H) Fusarium acuminatum. (I) F. avenaceum. (J) F. solani. (K) F. torulosum. (L) Mock-inoculated control. On each tap root, two to four inoculum plugs were placed in a line with same distance between each other. From the top and the tip of a root, the first and fourth plugs were directly placed on root surface and the second and the third plugs were placed on a punctured tissue, which was poked with an inoculation needle.

Fig. 7

Symptoms of red-skin root disease of Panax ginseng roots inoculated with the fungi under greenhouse conditions. (A) Dactylonectria sp. (B) D. hordeicola. (C) Ilyonectria changbaiensis. (D) I. communis. (E) I. mors-panacis. (F) I. qitaiheensis. (G) I. robusta. (H) Fusarium acuminatum. (I) F. avenaceum. (J) F. solani. (K) F. torulosum. (L) sterilized water.

Discussion

By analyzing 230 fungal isolates, we have determined that Asian ginseng red-skin root disease was caused by a complex of fungi, which consisted of 12 species. These fungi are all weak pathogens, which only resulted in red-skin root symptoms under greenhouse condition. Even though ginseng roots were acupunctured before inoculation in vitro, the disease lesions were around the inoculated site without further expanding.

Root diseases of ginseng are mainly attributed to Cylindrocarpon destructans [14], [37], the teleomorph of which is Ilyonectria spp. Most of them are soil inhabitants [19], [36], [38], [39], [40], [41]. However, the limited number of C. destructans isolates from Panax spp. was deduced into I. crassa, I. robusta, I. panacis, and I. mors-panacis [19]. We have found that Cylindrocarpon-like isolates were the most frequent organisms causing root disease in ginseng, and they belonged to 7 species in 2 genera: D. hordeicola, Dactylonectria sp., I. mors-panacis, I. robusta, I. changbaiensis, I. communis, and I. qitaiheensis. Dactylonectria hordeicola was described as Cylindrocarpon obtusisporum previously [42], which caused rusty root rot disease of Asian ginseng in China and showed weak virulence [16]. As red-skin disease and rusty root rot disease of Asian ginseng in China had causal pathogens in common, we suggest treating red-skin disease as rusty root rot at early stage of Asian ginseng.

Ilyonectria robusta was isolated from P. ginseng for the first time recently in China but was widely distributed at a high frequency [43]. It has a broad host range, including herbaceous plants Loroglossum hircinum and P. quinquefolium and woody plants Vitis vinifera, Prunus cerasus, Thymus sp., Quercus spp., and Tilia petiolaris [19]. Ramularia mors-panacis, Cylindrocarpon panacis, and Cylindrocarpon destructans f. sp. panacis were the basionyms of Ilyonectria mors-panacis [19], and that was reported to be the strong pathogenic species causing root rot disease on P. quinquefolium and P. ginseng [44], [45], [46]. Similarly, the only one isolate of I. mors-panacis we obtained did show a higher virulence compared with other Cylindrocarpon-like species under greenhouse conditions.

Ilyonectria crassa and I. panacis have been isolated from American ginseng in Canada [19]. We did not find I. crassa and I. panacis, but their sister species I. communis was new and named. Ilyonectria communis is characterized by branched conidiophores with up to four phialides, faster mycelial growth on PDA at 22°C in the dark and chlamydospores formed in the cells of microconidia, which can be clearly distinguished from the group I. pseudodestructans, I. crassa, I. rufa, and I. panacis. Ilyonectria changbaiensis and I. qitaiheensis were named by the only county where the isolates were collected from. Ilyonectria changbaiensis can be distinctly distinguished on frequently branched conidiophores with up to three phialides or wider 3-septate macroconidia, from the cluster I. qitaiheensis, I. gamsii, and I. liliigena. Ilyonectria qitaiheensis was characterized by faster mycelial growth on PDA at 22°C in the dark, longer 3-septate macroconidia and chlamydospores formed in the cells of microconidia. So far, the sister species I. gamsii and I. liliigena have not been isolated from Panax species [19]. Besides these Ilyonectria species, I. leucospermi was obtained from Korean ginseng roots recently [46], but we did not isolate I. leucospermi in this study.

Following Ilyonectria, Fusarium was the second most frequently isolated genus causing red-skin root disease on Asian ginseng. Among them, F. solani took 74.3% of the isolates. The rest of Fusarium isolates were F. acuminatum, F. avenaceum, and F. torulosum. Contrastingly, F. cerealis, F. redolens, and F. acuminatum have been reported to cause Asian ginseng root rot [47], [48], [49]. In this study, F. avenaceum and F. torulosum caused typical root rot symptoms on detached roots but caused red-skin symptoms after a growth season after inoculation under greenhouse condition. And, F. acuminatum caused both red-skin and root rot disease symptoms under greenhouse condition. Probably, F. acuminatum, F. avenaceum, and F. torulosum could cause either red-skin disease or root rot depending on the environmental conditions. Similar results have been reported in I. mors-panacis, which could cause root softening and also discoloration on Korean ginseng [46]. We suspect this may apply to other Cylindrocarpon-like species on Asian ginseng.

Among the Fusarium spp. causing red-skin root disease on Asian ginseng, F. avenaceum is also a causal agent of rusty root in American ginseng, but F. acuminatum F. solani and F. torulosum did not cause disease on American ginseng [20], [21]. Besides F. avenaceum, F. equiseti, F. sporotrichioides, and F. culmorum could infect American ginseng, and F. equiseti was a predominant pathogen causing discolored American ginseng roots [20], [21]. These results suggested that the predominant Fusarium species causing root disease of Asian ginseng in China were distinctive from those on American ginseng in North America. Whether the cause of differences is attributed to host or geography remained to be confirmed in our ongoing work.

Besides Cylindrocarpon-like and Fusarium species, several other species were isolated from symptomatic ginseng roots, such as Plectosphaerella cucumerina, Phoma exigua, Mortierella sp. and Rhexocercosporidium panacis. However, only R. panacis caused red-skin root symptoms [34], and it is not clear whether these isolates were pathogens and how they contributed to the symptom development. The clarification that the red-skin root of Asian ginseng is an infectious disease caused by several weak pathogenic fungal species will help develop disease management strategies.

Conflicts of interest

The authors have no conflicts of interest to report.