Thaumatin-like Protein (TLP) Genes in Garlic (Allium sativum L.): Genome-Wide Identification, Characterization, and Expression in Response to Fusarium proliferatum Infection.

Plant antifungal proteins include the pathogenesis-related (PR)-5 family of fungi- and other stress-responsive thaumatin-like proteins (TLPs). However, the information on the TLPs of garlic (Allium sativum L.), which is often infected with soil Fusarium fungi, is very limited. In the present study, we identified 32 TLP homologs in the A. sativum cv. Ershuizao genome, which may function in the defense against Fusarium attack. The promoters of A. sativumTLP (AsTLP) genes contained cis-acting elements associated with hormone signaling and response to various types of stress, including those caused by fungal pathogens and their elicitors. The expression of AsTLP genes in Fusarium-resistant and -susceptible garlic cultivars was differently regulated by F. proliferatum infection. Thus, in the roots the mRNA levels of AsTLP7-9 and 21 genes were increased in resistant and decreased in susceptible A. sativum cultivars, suggesting the involvement of these genes in the garlic response to F. proliferatum attack. Our results provide insights into the role of TLPs in garlic and may be useful for breeding programs to increase the resistance of Allium crops to Fusarium infections.

1. Introduction

Plant antifungal proteins provide resistance to fungal pathogens and are a focus of agricultural biotechnology. Overall, 13 classes of antifungal proteins are distinguished based on structural similarity and functional activity [1]. Among them, the pathogenesis-related (PR)-5 family members are homologous to the sweet-tasting protein thaumatin isolated from a Western African herb Thaumatococcus danielli Benth [2]. Thaumatin-like proteins (TLPs) are identified in more than 180 plants, including dicots, monocots, gymnosperms, bryophytes, and algae [3], and some of them display strong antifungal activity against Rhizoctonia solani, Alternaria alternata, Fusarium graminearum, Fusarium solani, Verticillium spp., and Phytophtora spp. [4,5,6]. TLPs could also be activated by bacterial pathogens, abiotic stresses (such as wounding, drought, osmotic stress, low temperature, high salinity, and UV radiation), and plant hormones [7,8,9,10,11,12,13,14,15].

Constitutive expression of TLP genes enhances plant tolerance to fungal pathogens [16,17,18,19,20,21,22,23,24]. Thus, transgenic tobacco plants overexpressing peanut, rice, or cotton TLPs show increased resistance to Botrytis cinerea, R. solani, Fusarium oxysporum, F. solani [23], A. alternata [17], and Verticillium dahliae [24], whereas transgenic potatoes overexpressing the Camellia sinensis TLP gene are resistant to Phytophthora infestans [19]. It has been reported that TLPs exert antifungal effects by lysing the fungal cell wall and inhibiting hyphal growth [25,26], which could be due to the β-glucanase activity of TLPs as well as to their role in the induction of pathogen defense-related mechanisms, including phenylpropanoid and phytoalexin production [10,18,24].

Interestingly, TLPs can also be expressed by plant pathogens as a part of their molecular mimicry strategy to evade the plant’s defense system and promote colonization of the host [27]. For example, TLPs of the nematode Bursaphelenchus xylophilus (Bx-TH1 and Bx-TH2) transiently expressed in Nicotiana benthamiana induce plant cell death [28,29].

TLPs are also known as pollen and fruit allergens, which stimulate IgE production, thus predisposing people to allergy [30,31,32].

As a group, TLPs represent highly soluble low molecular weight proteins classified into L-type (22–26 kDa) and S-type (<18 kDa) [33,34]; according to the isoelectric point (pI), some of them are highly acidic (pI = 3.4), whereas the other—very basic (pI = 12). TLPs are resistant to proteases and heat/pH-induced denaturation because of high stability of their structure due to disulfide bonds formed by 10 (S-type) or 16 (L-type) conserved cysteines [32]. Some plant species express TLPs fused with protein kinases (PR5- or TLP-kinases), which are suggested to serve as receptor protein kinases for pathogen sensing and activation of the downstream signaling [35,36,37,38,39]. Other members of the TLP family are osmotins associated with osmotic regulation, salt stress resistance, and antifungal activity [3,40] and permatins involved in antifungal defense of developing seeds [41,42].

In total, 344 plant TLPs have been identified through analysis of NCBI, EMBL, DDBJ, UniProt, and other databases [43]. Most of them contain an N-terminal signal peptide, specific glycoside hydrolase domain, highly conserved molecular signature motif GX[GF]XCXT[GA]DCX(1,2)GX(2,3)C, REDDD motif, and conserved Cys residues [33]. TLPs have endo-β-1,3-glucanase activity and can hydrolyze β-1,3-glucan of the fungal cell wall by binding to it through a negatively charged acidic interdomain cleft containing conserved residues underlying substrate binding (Lys and two Thr forming hydrogen bonds with (1,3)-β-d-glucan) and cleft acidity (Glu and three Asp) [27,44]. The acidic cleft and a significantly shortened so-called “thaumatin loop” of TLPs are responsible for the disappearance of sweet taste inherent to thaumatin, which has a basic cleft (surrounded with Lys residues replaced in TLPs by neutral Thr, Leu, and Ala and acidic Asp) [27].

The aim of this study was to characterize TLP genes in garlic (Allium sativum L.), one of the important bulb crops highly susceptible to infection with F. oxysporum f. sp. cepae (Fusarium basal rot, FBR) and F. proliferatum (bulb rot), which are responsible for more than 60% of the world’s garlic production losses at both pre- and post-harvest stages [45,46,47,48]. The disease symptoms include dry brown necrotic spots, white mycelium, and water-soaked signs at the clove surface [47]. A previous study indicates that the A. sativum PR5 (AsPR5) gene (AKU38392.1), which is upregulated by plant hormones methyl jasmonate (MeJA), abscisic acid (ABA), and ethylene, is also induced in response to F. oxysporum [49]. Given that in Arabidopsis, PR5 expression confers resistance to a necrotrophic fungus B. cinerea, it can be suggested that AsPR5 plays a similar role in garlic through regulation of signaling pathways associated with antifungal defense. The other three identified garlic TLP-coding genes, AsPR5a (Asa2G01043.1), AsPR5b (Asa4G02099.1), and AsPR5c (Asa4G02100.1) have been reported to be differentially expressed in the roots of FBR-resistant and -susceptible garlic cultivars in response to F. proliferatum attack [50].

Here, we identified in silico 32 putative AsTLP genes in the genome of A. sativum cultivar (cv.) Ershuizao, performed their structural characterization, analyzed phylogeny, and cloned AsTLP homologs from FBR-resistant and -susceptible garlic cultivars. Comparative expression profiling of AsTLP1–32 in various tissues of these cultivars infected with F. proliferatum suggests a role of AsTLPs in the mechanisms underlying garlic defense against fungal pathogens. Our results provide new insights into the function of A. sativum TLPs, which can be used in breeding programs to increase the resistance to Fusarium in cultivated Allium spp.

2. Results

2.1. In Silico Genome-Wide Identification of TLP Genes in A. sativum cv. Ershuizao

A total of 32 complete TLP gene sequences were detected in the genome and transcriptome of A. sativum cv. Ershuizao (PRJNA606385, PRJNA607255) and denoted AsTLP1–32 (Table 1, Supplementary Table S1). Most of the genes (AsTLP3–20) are located on chromosome 2 and seven (AsTLP1, 2, and 21–25)—on chromosomes 1, 4, 6, 7, and 8 (Figure 1), whereas the other seven (AsTLP26–32) have been found in scaffolds and do not match any chromosomes (assembly Garlic.V2.fa; Table 1). The sizes of the identified genes range from 615 to 2572 bp and those of the coding sequences (CDSs)—from 615 to 972 bp (Table 1).

Table 1

Characteristics of the predicted TLP genes in the A. sativum cv. Ershuizao genome.

Gene	Genomic Location (Strand)	A. thaliana Homolog	Transcript ID in RNA-Seq Database	Size (bp)	Exons	CDS (bp)	Protein
							Size (aa)	MW (kDa)	pI	Signal Peptide	GH64Domain	GRAVY
AsTLP1	Chr1:491780133..491781147 (+)	Pathogenesis-related thaumatin superfamily proteinNP_001031792.1	Asa2G02997.1	1013	2	930	309	32.29	4.93	1–24	31–245	0.126
AsTLP2	Chr1:996536859..996536201 (−)	Osmotin 34NP_192902.1	Asa2G01042.1	660	1	660	219	23.46	5.52	1–20	23–219	−0.157
AsTLP3 (AsPR5a)	Chr2:821219552..821220214 (−)		Asa2G01043.1	663	1	663	220	23.46	4.71	1–21	24–220	−0.085
AsTLP4	Chr2:1584165307..1584166065 (−)	TLPAAD02499.1	Asa3G06018.1	759	1	759	252	25.92	4.79	1–27	30–251	0.069
AsTLP5	Chr2:1617805168..1617805788 (−)	Osmotin 34NP_192902.1	Asa4G02103.1Asa4G02102.1	621	1	621	206	21.89	4.76	1–26	29–206	−0.289
AsTLP6	Chr2:1617847011..1617847625 (−)			615	1	615	204	21.74	4.76	1–24	27–204	−0.263
AsTLP7	Chr2:1617850920..1617851543 (−)		Asa4G02101.1	624	1	624	207	22.02	4.77	1–27	30–207	−0.295
AsTLP8 (AsPR5b)	Chr2:1618362333..1618362995 (+)		Asa4G02099.1	663	1	663	220	23.50	4.71	1–21	24–220	−0.126
AsTLP9 (AsPR5c)	Chr2:1618377059..1618377721 (+)		Asa4G02100.1	663	1	663	220	23.59	4.74	1–21	24–220	−0.164
AsTLP10	Chr2:1618405507..1618406165 (−)		Asa3G05858.1	660	1	660	219	23.43	4.71	1–20	23–219	−0.126
AsTLP11	Chr2:1622416758..1622417381 (−)		Asa3G05840.1	624	1	624	207	22.08	5.70	1–27	30–207	−0.211
AsTLP12	Chr2:1684781475..1684782236 (+)	TLPAAD02499.1	Asa3G06302.1	762	1	762	253	26.16	4.27	1–29	32–252	0.056
AsTLP13	Chr2:1685003219..1685003978 (−)		Asa3G06331.1	759	1	759	252	26.28	4.24	1–28	31–251	0.012
AsTLP14	Chr2:1685049653..1685050414 (−)		no detected	762	1	762	253	26.06	4.18	1–29	32–252	0.087
AsTLP15	Chr2:1685078657..1685079418 (−)		no detected	762	1	762	253	26.06	4.18	1–29	32–252	0.068
AsTLP16	Chr2:1685746745..1685747579 (+)		Asa3G06310.1	835	2	747	248	25.30	4.59	1–25	27–247	0.031
AsTLP17	Chr2:1686634047..1686634808 (+)	Pathogenesis-related thaumatin superfamily proteinNP_173432.2	Asa3G06319.1	762	1	762	253	26.09	4.18	1–29	32–252	0.054
AsTLP18	Chr2:1687188296..1687189055 (−)	TLPAAD02499.1	Asa3G06331.1	759	1	759	252	26.30	4.24	1–28	31–251	0.013
AsTLP19	Chr2:1687209664..1687210422 (−)		Asa3G06329.1	759	1	759	252	26.21	4.27	1–28	31–251	0.025
AsTLP20	Chr2:1724629365..1724630114 (+)	Pathogenesis-related thaumatin superfamily proteinNP_973870.1	Asa7G02264.1	750	1	750	249	25.42	4.39	1–26	28–248	0.081
AsTLP21	Chr4:777618126..777620378 (+)	Pathogenesis-related thaumatin superfamily proteinNP_568046.1	Asa4G00901.1	2253	3	972	323	33.86	4.50	1–19	26–232	−0.008
AsTLP22	Chr6:2008642583..2008643344 (+)	TLPAAD02499.1	Asa3G06325.1	762	1	762	253	26.09	4.18	1–29	32–252	0.085
AsTLP23	Chr7:682376071..682376964 (+)	Pathogenesis-related thaumatin superfamily proteinNP_001119140.1	Asa5G05974.1	894	1	894	297	31.13	4.68	1–21	23–242	0.115
AsTLP24	Chr8:1265530915..1265533486 (−)	Pathogenesis-related thaumatin superfamily proteinNP_177503.1	Asa7G04908.1	2572	2	753	250	25.97	9.07	1–26	31–247	0.081
AsTLP25	Chr8:1394582992..1394583731 (+)	Pathogenesis-related thaumatin superfamily protein NP_001324474.1	Asa7G04378.1	741	1	741	246	26.03	8.10	1–21	24–245	−0.005
AsTLP26	Scaffold9088: 51512..52273	Pathogenesis-related thaumatin superfamily proteinNP_173432.2	Asa3G06323.1	762	1	762	253	26.09	4.18	1–29	32–252	0.054
AsTLP27	Scaffold9089: 39596..38835	TLPAAD02499.1	Asa3G06322.1	762	1	762	253	26.12	4.18	1–29	32–252	0.075
AsTLP28	Scaffold9089: 40346..41107			762	1	762	253	26.12	4.18	1–29	32–252	0.075
AsTLP29	Scaffold9089: 89114..89875		Asa3G06320.1	762	1	762	253	26.06	4.18	1–29	32–252	0.044
AsTLP30	Scaffold9091: 149728..150486		Asa3G06315.1	759	1	759	252	26.28	4.24	1–28	31–251	0.006
AsTLP31	Scaffold12619: 87332..88248		no detected	917	3	738	245	26.47	5.75	1–22	25–244	−0.134
AsTLP32	Scaffold12619: 99641..100558		no detected	918	3	738	245	26.46	5.75	1–22	25–244	−0.149

Figure 1

Locations of AsTLP genes in the A. sativum chromosomes. Chromosome lengths (indicated on the left) are based on the A. sativum cv. Ershuizao genome (PRJNA606385; assembly Garlic.V2.fa); chr, chromosome.

2.2. Structural and Phylogenetic Analyses of AsTLPs

Most identified AsTLP genes (26) did not have introns, and the rest contained one (AsTLP1, 16, 24) or two (AsTLP21, 31, 32) introns (Figure 2a, Table 1). The characteristics of the translated proteins are shown in Table 1. Putative AsTLPs ranged in size from 204 to 323 amino acids (aa); among them, 13 were hydrophilic and 19-hydrophobic according to hydrophobicity indicators. All AsTLPs had similar structures, containing an N-terminal signal peptide (21–29 aa), full-length functional catalytic domain GH64-TLP-SF (glycoside hydrolase family 64 and TLP superfamily; pfam00314), thaumatin signature GX[GF]XCXT[GA]DCX(1,2)GX(2,3)C (PS00316), “REDDD” motif associated with antifungal activity, and 13–18 cysteine residues responsible for disulfide bond formation [44] (Figure 2b and Figure 3a, Table 1).

Figure 2

Phylogenetic and structural analysis of AsTLPs. (a) Evolutionary relationship based on amino acid sequences and exon-intron structures of the AsTLP genes. The unrooted dendrogram was constructed in MEGA 7.0.26 using the Neighbor-Joining method (bootstrap test: 1000 replicates). (b) Sequence alignment of AsTLPs. Regions with 50–100% identity are grey-shaded. The conserved thaumatin domain (GH64-TLP-SF; pfam00314) is underlined blue, the TLP family signature GX[GF]XCXT[GA]DCX(1,2)GX(2,3)C (PS00316) is framed red, and REDDD motif residues are marked with asterisks.

Figure 3

Position of the thaumatin domain (violet box) (a) and distribution of conserved motifs (b) in AsTLPs. Analysis was performed using MEME 5.3.0. The length of each box corresponds to that of the motif.

Phylogenetic analysis revealed six AsTLP clades (Figure 2A). The first and largest clade comprised 16 AsTLPs (73.4–99.6% identity), the second contained AsTLP31 and AsTLP32 (98.8% identity), the third contained AsTLP1 and AsTLP23 (63.9% identity), the fourth contained AsTLP24 and AsTLP25 (42.0% identity), the fifth containedAsTLP5 –7 and AsTLP11 (92.3–97.6% identity), and the sixth contained AsTLP2, AsTLP3, and AsTLP8–10 (95.9–99.1% identity). In the NCBI database of non-redundant protein sequences, AsTLPs of the last two clades show strong homology to A. thaliana osmotin 34 (ID: NP_192902.1) (Table 1).

The search for allergenic motifs disclosed statistically significant sequence similarity between AsTLPs and thaumatin-like allergens from apple (Mal d 2) [51], peach (Pru p 2.0101 and Pru p 2.0201) [52], cherry (Pru av 2) [31], and kiwi (Act c 2) [53].

In total, 15 conserved motifs were identified in putative AsTLPs, including motif 8 corresponding to signal peptide. MEME analysis indicated that phylogenetically related AsTLPs belonging to the same clade had similar conserved motifs (Figure 3b). Thus, motif 15 (consensus RHWNPQGLVPAVGDGMIJF) was detected only in clade III, motif 13 (WSGRIWGRQGC)—in clade IV, and motif 12 (QLSCTGTGQTPATL)—in clade V. The differences observed in the motif composition of the thaumatin domain corresponded to protein phylogeny (Figure 3b).

Annotation of AsTLPs in Gene Ontology (GO) terms predicted their extracellular localization (GO:0005576), as well as a role in defense response (GO:0006952) and response to fungal infection (GO:0009620).

2.3. In Silico Analysis of AsTLP Transcription

The expression of AsTLP genes in various tissues (roots, bulbs, stems, leaves, buds, flowers, and sprouts) of A. sativum cv. Ershuizao was evaluated based on transcriptomics data (PRJNA607255). As the bulb is the part of garlic most susceptible to Fusarium infection and shows visible symptoms, gene expression analysis in this organ was more detailed and included eight developmental stages (from day 192 to day 227 after planting). Figure 4 shows mRNA expression levels of 16 AsTLP genes, for which Fragments Per Kilobase of transcript per Million mapped reads (FPKM) was ≥ 10 in at least one of the organs. AsTLP1–3, 5–11, and 16 had overall similar expression patterns (with the exception of AsTLP1 and 9 in the roots and AsTLP5-7, 9, 11, and 16 in the leaves), whereas those of AsTLP20, 21, and 23–25 were different.

Figure 4

Heatmap of AsTLP gene expression in A. sativum tissues. Gene transcription was analyzed in the roots, bulbs (1, 2, 3, 4, 5, 6, 7, and 8 correspond to 192-, 197-, 202-, 207-, 212-;, 217-, 222-, and 227-day-old bulbs), stems (ps.stem), leaves, buds, flowers, and sprouts. The color gradient indicates expression changes from low (red) to high (green).

About a third of the AsTLP genes (AsTLP1–3, 5–11, and 16) were strongly expressed in the stems and leaves. In the roots, the highest mRNA levels were observed for AsTLP1, 9, and 23 and the lowest—for AsTLP21, 24, and 25; the latter had the highest expression in the floral buds and stage 1 bulbs, which gradually decreased during bulb development. The expression of AsTLP20 and 23 in the bulbs showed bell-shape changes, with the maximum at stages 3 and 6, respectively. According to the expression profiles, the AsTLP genes could be divided into two groups: AsTLP1–3, 5–11, and 16 with the highest expression in the stems, leaves, and roots, and AsTLP20, 21, 24, and 25 with that in the developing bulbs, floral buds, and flowers. The first group includes all genes of phylogenetic clades V and VI, as well as one gene each from clades I and III; members of the second group are phylogenetically heterogeneous (Figure 2a). The expression pattern of the AsTLP23 gene had the features of both groups (Figure 4).

2.4. AsTLP Promoter Analysis

Considering the stress- and hormone-responding role of TLPs [49], we searched for respective cis-acting elements in the 5′-UTR and promoter regions (1 kb upstream of the initiation codon) of AsTLP genes. The results indicated that the AsTLP regulatory regions contained 11 hormone-responsive and 10 stress-responsive elements (Figure 5). Among the former, the most common were the ABA-responsive element (ABRE) [54], ethylene-responsive element (ERE) critical for ethylene-regulated transcription in plants [55], a highly conserved prolamin-box (P-box) recognized by the trans-activator P-box-binding factor (PBF) during endosperm development [56], and the CGTCA motif involved in MeJA and osmotic stress responsiveness [57]. Among the stress-responsive elements, the most common were anaerobic-responsive element (ARE) involved in the activation of anaerobic gene expression [58] and TC-rich repeats found in promoters of many plant disease-resistance genes [59]. The promoters most enriched in the regulatory elements were those of AsTLP11 (2 ABRE, 4 CGTCA motifs, 2 ERE, and 3 ARE), AsTLP17 (4 ERE and 3 ARE), AsTLP1, 24, and 25 (7, 4, and 3 ABRE, respectively), and AsTLP6 and 10 (4 and 5 ARE, respectively). The promoters of five genes (AsTLP8, 9, 23, 24, and 32) contained low-temperature-responsive (LTR) elements; among them, AsTLP23 had the highest number (6). The W-box, which responds to fungal elicitors and wounding, was found in only three genes (AsTLP1, 24, and 30). The rarest elements were ABA-responsive CARE (in AsTLP1 and 21) and auxin-responsive AuxRE (in AsTLP8 and 9) (Figure 5). No pronounced similarity between the set of elements and the structure of genes was found, with the exception of some groups of genes, for example, AsTLP5 and 6 (clade V), 8 and 9 (clade VI), or 22, 28, and 29 (clade I).

Figure 5

Hormone- and stress-responsive cis-elements in the regulatory regions (~1000 bp) of AsTLP genes. The color scheme (from pale to dark) corresponds to the numbers of cis-elements (from low to high).

2.5. AsTLP1-32 Expression in cv. Sarmat and Strelets Infected with F. proliferatum

To elucidate the role of the identified AsTLP1-32 genes in the garlic response to fungal infection, we performed AsTLP expression profiling in garlic cultivars resistant (cv. Sarmat) and susceptible (cv. Strelets) to FBR [60]. The AsTLP7 corresponds to clade V group of genes (AsTLP5–7 and 11), AsTLP8—to set of AsTLP3, 8, and 10 genes, AsTLP12—to set of AsTLP12, 15, 17, 22, and 26-29 genes, AsTLP19—to AsTLP13, 18, 19, and 30 genes, and AsTLP31—to AsTLP31 and 32 (clade II) genes (Supplementary Table S2). Gene expression was analyzed in the roots, stems (basal plates), and cloves of cv. Sarmat and cv. Strelets at 24 and 96 h post infection (hpi) with F. proliferatum and compared with uninfected control (Figure 6 and Figure 7).

Figure 6

Expression of the AsTLP genes in the roots, stems, and cloves of A. sativum FBR-resistant cv. Sarmat and FBR-susceptible cv. Strelets at 24 and 96 hpi with F. proliferatum. AsTLP8 represents the cumulative data for AsTLP3, 8 and 10, AsTLP12—for AsTLP12, 15, 17, 22, and 26–29, and AsTLP19—for AsTLP13, 18, 19, and 30. Transcription of AsTLP5, 6, and 32 was not detected. The data were normalized to GAPDH and UBQ mRNA levels and presented as the mean ± SE (n = 3) compared to control (cv. Sarmat expression at 24 hpi taken as 1); * p < 0.01 compared to uninfected control.

Figure 7

Time-dependent expression of AsTLP genes in the roots, stems (basal plates), and cloves of uninfected garlic cultivars resistant (cv. Sarmat) and susceptible (cv. Strelets) to FBR. The data were normalized to GAPDH and UBQ mRNA levels and presented as the mean ± SE (n = 3); * p < 0.01 indicates the difference between 24 and 96 h.

The results revealed that the AsTLP genes were transcribed in all analyzed organs, except for AsTLP24 not expressed in the cloves (Figure 6). To determine the expression of individual genes in clades V and II, we performed sequencing of PCR-amplified products, which revealed that only AsTLP7 and AsTLP31 were transcribed in the tissues of both cultivars. For the other groups of highly homologous AsTLPs, the amplified PCR products were invariable within each group.

In both uninfected cultivars, the highest expression was observed for AsTLP7, 8, and 31 (Figure 7). Analysis of time-dependent changes revealed that most genes had similar expression dynamics in the two cultivars. Thus, the mRNA levels of AsTLP1, 7, 8, and 16 remained the same or increased from 24 to 96 hpi, whereas those of AsTLP12, 19–21, and 24 were either the same or decreased. However, the expression of AsTLP23 and 31 was increased from 24 to 96 hpi in the cloves of FBR-resistant cv. Sarmat but decreased in those of FBR-susceptible cv. Strelets (Figure 7).

Symptoms of F. proliferatum infection were observed only in FBR-sensitive cv. Strelets after 96 hpi: the roots were covered with white fungal mycelium, indicating the external growth of the pathogen. Infection changed the expression patterns of the AsTLP genes compared to uninfected control. Thus, in FBR-resistant cv. Sarmat, the transcription of all genes was upregulated at 24 hpi, except for that of AsTLP1 in the roots and stems, which was upregulated at 96 hpi. These data suggest a correlation between the expression level of AsTLP1 and the appearance of disease symptoms. In the roots, the expression of most genes was increased from 24 to 96 hpi, except for that of AsTLP12 and 19 (unchanged) and AsTLP24 (decreased). Similar dynamics were observed in the cloves, except for AsTLP12 and 31, which were downregulated with time. In the stems, the expression of AsTLP7–9 and 31 decreased and that of AsTLP19 and 24 increased from 24 to 96 hpi, whereas that of AsTLP16 was unchanged (Figure 6).

In FBR-susceptible cv. Strelets, all AsTLP genes were upregulated at 24 hpi compared to uninfected control, except for AsTLP9 in the roots and stem and AsTLP31 in the cloves, which were upregulated at 96 hpi (Figure 6). The time-dependent AsTLP expression dynamics in cv. Strelets significantly differed from that in cv. Sarmat: the expression of AsTLP7–9, 20, 21, 23, and 24 was either unchanged or downregulated from 24 to 96 hpi, whereas tissue-dependent changes were observed for the other genes. Thus, AsTLP12 and 31 were downregulated with time in the roots and cloves and upregulated in the stems, AsTLP16 was downregulated in the roots and stems and upregulated in the cloves, and AsTLP19 was downregulated in the cloves and upregulated in the stems (Figure 6).

2.6. Cloning and Characterization of CDSs of AsTLP Genes Differentially Expressed in FBR-Sensitive and -Resistant Cultivars

Considering the differential expression of some AsTLP genes in cv. Sarmat and cv. Strelets in response to F. proliferatum infection (Figure 6), we investigated the association of these genes with FBR resistance. AsTLP7, 16, and 20 were upregulated by the infection in the roots of cv. Sarmat, whereas AsTLP21 and 23 were upregulated in those of cv. Strelets. The CDSs of these genes were amplified, cloned, and sequenced, and the data deposited in NCBI GenBank (Table 2).

Table 2

Polymorphisms in the AsTLP CDSs of cv. Sarmat and Strelets compared to cv. Ershuizao.

Gene	NCBI IDcv. Sarmat/cv. Strelets	cv. Sarmat	cv. Strelets
		SNPs (aa Substitution)
AsTLP7	OM386716/OM386717	c.183A>C, c.218G>C (p.G73A), c.219A>C (p.G73A), c.225A>C, c.233A>G (p.N78S), c.243G>C, c.249G>A, c.252T>C, c.276T>C, c.279T>C, c.303A>C, c.333T>C, c.357T>C, c.378C>G, c.384C>T, c.399 >T, c.479>T (p.T160M), c.615C>T
AsTLP16	OM386718/OM386719	c.141T>C, c.162T>C, c.406C>T, c.476C>G (p. A159G)
AsTLP20	OM386720/OM386721	c.49T>C (p.S17P), c.53G>T (p.C18F), c.67T>C (p.S23P), c.70A>G, c.75A>T, c.184G>C (p.G62R)
			c. 670T>C (p.Y224H)
AsTLP21	OM386722/OM386723	c.22A>G, c.141G>A, c.453A>G, c.554G>A, c.570C>T, c.573T>G, c.738A>G, c.804A>T, c.929C>G (p.A310G)
AsTLP23	OM386724/OM386725	c.T30>C, c.A45>T (p.Q15H), c.A99>G, c.T147>G, c.C165>G, c.C588>G, c.C684>T, c.C821>G (p. A274G), c.C847>T, c.C849>T, c.T851>C

Note: Non-synonymous SNPs and corresponding amino acid substitutions are marked in bold.

The cv. Ershuizao AsTLP sequences were used as references. The amino acid sequences of AsTLP7, 16, 20, 21, and 23 from cv. Sarmat and Strelets did not differ in size from those of cv. Ershuizao AsTLPs. Compared to the cv. Ershuizao AsTLP genes, the cv. Sarmat and Strelets genes contained 4–18 single nucleotide polymorphisms (SNPs). Among them, non-synonymous SNPs leading to amino acid substitutions were found in AsTLP7 (4), AsTLP16 (1, resulting in radical A159G), AsTLP20 (4, including radical G62R), AsTLP21 (1), and AsTLP23 (1) of both cultivars, whereas a cultivar-specific non-synonymous SNP (670T>C, leading to radical Y224H) was detected in AsTLP20 of cv. Strelets (Table 2).

2.7. Analysis of Regulatory Regions in the AsTLP Genes Differentially Expressed in cv. Sarmat and Strelets after F. proliferatum Infection

The regulatory sequences (promoter and 5′-UTR) of the AsTLP7, 16, 20, 21, and 23 genes, which were differently expressed in cv. Sarmat and Strelets in response to F. proliferatum infection, were amplified, sequenced, and analyzed for hormone- and stress-responsive elements (Figure 8). The regulatory region of only the AsTLP23 gene of cv. Sarmat and Strelets differed from the AsTLP23 of cv. Ershuizao by the absence of the CGTCA-motif and one of the six LTR-elements (Figure 8). In general, the sequences of the regulatory regions in AsTLP7, 16, 20, 21, and 23 were the same in cv. Sarmat and Strelets, but differed from those in cv. Ershuizao: there were 3 indels (11, 4, and 2 bp) and 20 SNPs in AsTLP7, 8 SNPs in AsTLP20, 3 indels (2, 2, and 7 bp) and 7 SNPs in AsTLP21, and 5 SNPs in AsTLP23.

Figure 8

Comparative analysis of hormone- and stress-related cis-elements in the promoter regions of AsTLP7, AsTLP16, AsTLP20, AsTLP21, and AsTLP23 genes in A. sativum cv. Ershuizao, cv. Sarmat, and cv. Strelets. The numbers of cis-elements are indicated. The elements present in cv. Sarmat and cv. Strelets but absent in cv. Ershuizao are highlighted green and those present in cv. Ershuizao but absent or less abundant in cv. Sarmat and cv. Strelets are highlighted red.

The mutations found in the AsTLP7, 20, and 21 regulatory sequences did not affect cis-elements, whereas those in the AsTLP23 sequences did. Thus, two SNPs (at positions -674G>A and -676A>G) resulted in the loss of the CGTCA motif associated with MeJA response, another SNP (-521C>T) created the GARE motif (gibberellin-responsive element), and two SNPs (-640G>A and -641C>A) eliminated an LTR element (Figure 8).

3. Discussion

Garlic (A. sativum L.) belongs to the Allium genus, which comprises 971 species, thus being one of the largest genera of the Amaryllidaceae family widely distributed in the Northern Hemisphere from dry subtropics to boreal zones [61,62]. Significant garlic crop losses regularly occur worldwide, mostly because of diseases caused by soil fungi of Fusarium spp. [63,64], as well as by drought, osmotic, and cold stresses [35,65]. In many plants, TLPs play a critical role in the defense against both abiotic and biotic stresses [66], suggesting that they also perform a similar role in garlic and may potentially be used in Allium crop breeding programs.

In the present study, we identified and characterized 32 TLP genes in the A. sativum cv. Ershuizao genome [67] and amplified five of them (CDSs and 5′-regulatory regions) from garlic cv. Sarmat and Strelets differing in the susceptibility to FBR. All the genes encoded proteins homologous to the A. thaliana pathogenesis-related TLPs, including osmotin 34, implying the same functional activity. Among the identified AsTLP genes, eight were located in six chromosomes, where they probably have emerged because of local segmental duplications, and 17 were tandemly arranged in a single cluster in chromosome 2 (Figure 1), which may be due to tandem duplication through multiple episodes of unequal crossovers. This hypothesis is consistent with the clade distribution of AsTLP genes, when several tandemly duplicated genes were clustered together (Figure 2a), suggesting their origin from recent duplication events. Structure analysis revealed that most AsTLPs (26) had no introns or contained only a few (1–2) of them (Figure 2a). The intron-poor and intronless genes transcribed with less or without splicing are considered to have originated from intron-rich genes as a part of the adaptation strategy aimed to provide fast responses to different biotic and abiotic stresses, which is consistent with strong induction of intronless genes after stress [68].

All putative AsTLP1–32 proteins contained the conserved thaumatin domain and TLP family motif signatures (Figure 2b and Figure 3), suggesting conservation of garlic TLPs in terms of regulation through stress/defense-related signaling and functional activity to hydrolyze fungal β-1,3-glucan. The presence of N-terminal signal peptides indicates targeting of mature AsTLPs to the apoplastic pathway, which is supported by GO analysis predicting AsTLP extracellular secretion and is consistent with a previous report on TLP localization to the intercellular space [69]. These data agree well with the function of TLPs in hydrolyzing the fungal cell wall and their GO-predicted role in defense response to fungal infection.

All AsTLP1–32 proteins contained fragments with significant homology to previously described thaumatin-like allergens from various plant species [31,51,52,53]. Given the reports of IgE-mediated human hypersensitivity to raw garlic [70], it may be attributed, at least in part, to the AsTLPs expressed in the cloves (Figure 6).

Seven AsTLPs (AsTLP5–11) were found to be homologous to A. thaliana osmotin 34 (Table 1), which functions in the initiation of ABA responses, including the regulation of ABA-induced proline synthesis [10], suggesting the involvement of AsTLP5–11 in the garlic response to osmotic stress.

It is known that fungi-mediated biotic stress activates plant immune system through perception of pathogen-associated molecular patterns [71] with following formation of reactive oxygen species, induction of salicylic acid and jasmonic acid signaling, and upregulation of PR genes [72]. The promoters of all PR5 family AsTLP genes contained cis-regulatory elements associated with the activation of immune mechanisms through responses to stresses (anaerobic conditions, dehydration, low temperature, salinization, heavy metals, and wounding), elicitors, and hormones, (Figure 6). A similar set of elements has also been found in the promoters of Gossypium barbadense and Rosa chinensis TLP genes [73,74] and of the other garlic PR1–5 genes [50,60]. Accordingly, the expression of AsTLP genes may be triggered by phytohormones (ethylene, salicylic acid, ABA, and MeJA), and stresses.

Ethylene and ABA homeostasis is known to be modulated in host plant in the response to fungal infection [60,75]. In accordance with this, the ERE and ABRE elements were found in the promoters of more than half of AsTLPs (Figure 5), which suggests these genes activation by ethylene and ABA after Fusarium attack.

Ethylene is known to crosstalk with ABA in response to abiotic stresses [75], such as low temperature (freezing), drought, and salinity, which cause osmotic stress and, consequently, activation of SnRK2 kinases as key signal transducers in the ABA pathway [76]. Considering the set of discovered stress responsive elements, it can be assumed that almost all AsTLP genes are involved in the response to osmotic stress, except for AsTLP7, 9, 16, and 20, promoters of which do not contain neither the CGTCA motif, nor the DRE1, LTR, and MBS elements. Only five genes (AsTLP8, 9, 23, 24, and 32) may be involved in the response to cold stress (especially AsTLP23, which contains 5 LTR elements) (Figure 5).

Some AsTLP genes within the same clade (AsTLP5–7, 11 or AsTLP2, 3, 8–10) (Figure 2a) tended to be dissimilar in the promoter cis-regulatory motif patterns (Figure 5), suggesting that their transcription in response to adverse conditions might be differentially regulated.

Analysis of TLP genes with the established function in plant defense has predicted two unique combinations of regulatory elements to distinguish between abiotic and biotic stress responses: genes regulated by abiotic stresses contain the ABRE element, whereas those responding to fungal attack do not [77]. These data suggest that AsTLP3, 4, 7, 14–17, 20–23, 26–29, and 31, whose promoters do not have ABRE (Figure 5 and Figure 8), may be involved in the antifungal defense of garlic.

Among the five AsTLP genes amplified from FBR-resistant cv. Sarmat and FBR-susceptible cv. Strelets, only one contained a cultivar-specific SNP leading to an amino acid substitution (Table 2), suggesting that the antifungal resistance in garlic may not be attributed to AsTLP mutations, but rather to transcriptional and translational regulation of gene expression. This notion is supported by the differential expression of AsTLP genes in FBR-resistant and -susceptible cultivars infected with F. proliferatum. The most pronounced dissimilarity was the time-dependent upregulation of AsTLP7–9 (homologs of OSMOTIN 34) and AsTLP21 in the roots of cv. Sarmat and their downregulation in those of cv. Strelets (Figure 6), suggesting that the expression dynamics of these AsTLP genes may underlie the difference in antifungal resistance between the cultivars. Considering that the promoters of AsTLP8 and 9 contain the ABRE element, whereas those of AsTLP7 and 21 do not, the latter are more likely to have a role in the garlic defense responses against F. proliferatum.

It has been suggested that TLP genes can serve as molecular markers associated with resistance to fungal diseases [69]. Thus, our results could be useful for breeding programs aimed on increasing the resistance of garlic crops to Fusarium spp. by, for example, generating cultivars that overexpress the respective AsTLP genes, which may protect against fungal infections. In addition, it should be noted that, besides TLPs, the fungicidal effects of Allium plants are associated with chitinases and endo-1,3-β-glucanases, as well as miRNAs involved in positive (miR164a, miR168a, and miR393) and negative (miR394) regulation of resistance to Fusarium [48,50,60,78]. Moreover, Allium roots produce antifungal volatiles such as 2-methyl-2-pentenal and organosulfur compounds, as well as non-volatiles such as spirostanol, furostanol, and steroidal saponins, which inhibit Fusarium growth [60,79,80].

4. Materials and Methods

4.1. In Silico Identification and Structural Characterization of AsTLP Genes

The search for TLP genes was performed in the A. sativum cv. Ershuizao whole-genome (NCBI accession number: PRJNA606385, assembly Garlic.V2.fa) and transcriptome (PRJNA607255) sequences [68]. The thaumatin domain (http://pfam.xfam.org/family/PF00314; accessed on 25 October 2021) was used as reference. All the selected sequences contained start and stop codons and full-length catalytic domains.

Multiple sequence alignment and structural analyses of the AsTLP genes and encoded proteins were conducted with MEGA 7.0.26 [81]. The phylogenetic dendrogram was constructed based on protein sequences using the MEGA 7.0.26 (ML method); confidence for tree topologies was estimated by bootstrap values of 1000 replicates.

To predict exon–intron composition, AsTLP genes and CDSs were analyzed with GSDS v2.0 [82]. Putative proteins were characterized by molecular weight, pI, and grand average hydropathy (GRAVY) (ExPASy ProtParam; https://web.expasy.org/protparam/; accessed on 30 October 2021; GRAVY indexes were calculated as the sum of hydrophobicity values of all residues divided by sequence length), conserved domains, sites, and motifs (NCBI-CDD, https://www.ncbi.nlm.nih.gov/cdd; accessed on 30 October 2021; and Multiple Em for Motif Elicitation (MEME 5.3.0), http://meme-suite.org/tools/meme; accessed on 30 October 2021), biological processes (PANNZER2; http://ekhidna2.biocenter.helsinki.fi/sanspanz/; accessed on 30 October 2021), subcellular localization (BaCello; http://gpcr2.biocomp.unibo.it/; accessed on 30 October 2021), functional importance of residue substitutions (PROVEAN; http://provean.jcvi.org/seq_submit.php; accessed on 25 November 2021), and signal peptide cleavage sites (SignalP 5.0; http://www.cbs.dtu.dk/services/SignalP/; accessed on 30 October 2021). The chromosomal localization map was drawn using MG2C v. 2.1 (http://mg2c.iask.in/mg2c_v2.1/; accessed on 30 October 2021). The search for allergenic motifs in AsTLPs was conducted using the Structural Database of Allergenic Proteins (SDAP) (https://fermi.utmb.edu/SDAP/sdap_fas.html; accessed on 19 January 2022).

4.2. In Silico mRNA Expression Analysis

The expression of AsTLP genes in the roots, bulbs, stems (basal plates), leaves, buds, flowers, and sprouts was analyzed based on A. sativum cv. Ershuizao RNA-Seq data (FPKM; ID: PRJNA607255), normalized as FPKM [66], and visualized using Heatmapper [83]. Only transcripts with an average FPKM value ≥ 10 in at least one of the organs were used for heatmap construction.

4.3. Gene Identification

To amplify the AsTLP CDSs from garlic cultivars, gene-specific primers were designed based on A. sativum cv. Ershuizao transcriptomic data (NCBI project accession number: PRJNA607255) (Supplementary Table S2). CDNA (30 ng) from the roots of each cultivar accession was used as a template for PCR amplification at the following conditions: initial denaturation at 95 °C for 5 min, 35 cycles [denaturation at 95 °C for 30 s, primer annealing at 55 °C for 30 s, and extension at 72 °C for 2 min], and final extension at 72 °C for 5 min. PCR products of the expected size were purified by using the QIAEX® II Gel Extraction kit (QIAGEN, Hilden, Germany), cloned in the pGEM®-T Easy vector (Promega, Madison, WI, USA), and sequenced (3–5 clones for each accession) on ABI Prism 3730 DNA Sequencer (Applied Biosystems, Waltham, MA, USA) using the designed primers.

4.4. Plants, Fungi, and F. proliferatum Infection Assay

F. proliferatum was kindly provided by the Group of Experimental Mycology, Winogradsky Institute of Microbiology (Research Center of Biotechnology of the RAS, Moscow, Russia). The strain was originally isolated from the bulbs of garlic cv. Strelets; according to the pathogenicity test, the first signs of the disease appeared on the clove surface 5 days after infection [50].

Accessions of A. sativum cv. Sarmat and cv. Strelets (winter garlic of Russian breeding) resistant and susceptible to FBR, respectively, were kindly provided by the Federal Scientific Vegetable Center (Moscow region, Russia). The number of cloves used per biological replicate in the Fusarium infection assay was based on that of cloves in the bulb (5–7 for cv. Strelets and 7–11 for cv. Sarmat). In total, 12 cloves of each cultivar (6 infected and 6 uninfected) were processed (three biological replicates were used). Cloves were surface-sterilized in 70% ethanol for 3 min, rinsed with sterile water, placed in Petri dishes with wet filter paper, and incubated at +25 °C in the dark. After 72 h, active root growth was observed, and half of the cloves were infected by soaking in F. proliferatum conidial suspension (~106 conidia ml−1) for 5 min as previously described [60]), transferred to fresh Petri dishes, and incubated at +25 °C in the dark for 24 and 96 h (n = 3 cloves per each time point).The roots, stems, and cloves of the infected and uninfected samples were collected at each time point, frozen in liquid nitrogen, and stored at −80 °C. The time points were chosen according to the expression peaks of some PR genes, which were observed 1–3 days after inoculation with hemibiotrophic pathogens [84].

4.5. RNA Extraction and Quantitative Real-Time Reverse Transcription PCR (qRT-PCR)

Total RNA was extracted from individual roots, stems, and cloves (0.5 g of each tissue) using the RNeasy Plant Mini Kit (QIAGEN, Hilden, Germany), purified from genomic DNA (RNase free DNase set; QIAGEN), qualified by gel electrophoresis, and used for first-strand cDNA synthesis (GoScript Reverse Transcription System; Promega, Madison, USA) with an oligo-dT primer. RNA and cDNA concentrations were quantified by fluorimetry (Qubit® Fluorometer, Thermo Fisher Scientific, Waltham, MA, USA) and qRT-PCR was performed in a CFX96 Real-Time PCR Detection System (Bio-Rad Laboratories, Hercules, CA, USA) with 3.0 ng cDNA, SYBR Green RT-PCR mixture (Syntol, Moscow, Russia), and specific primers (Supplementary Table S2). Because CDSs of some AsTLPs had a high degree of homology, universal primers were designed for such genes, which were grouped according to their homology (Supplementary Table S2). The following cycling conditions were used: initial denaturation at 95 °C for 5 min, 40 cycles of denaturation at 95 °C for 15 s, and annealing/extension at 60 °C for 40 s.

AsTLP gene expression was normalized using two reference garlic genes, GAPDH and UBQ [60], and the qRT-PCR results were statistically analyzed with Graph Pad Prism version 8 (GraphPad Software Inc., San Diego, CA, USA; https://www.graphpad.com/scientific-software/prism/ (accessed on 30 October 2021)). The data were expressed as the mean ± standard deviation (SE) based on three technical replicates of three biological replicates for each combination of cDNA and primer pairs. The unequal variance (Welch’s) t-test was applied to assess differences in gene expression; p < 0.01 was considered to indicate statistical significance.

4.6. Promoter and 5′-UTR Analysis

The search for specific cis-elements in the promoters and 5′-UTRs (1.0 kb regions upstream of the initiation codon) was performed using the PlantCARE database, which provides evaluation of cis-regulatory elements, enhancers, and repressors; (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/; accessed on 25 November 2021).

5. Conclusions

We identified and characterized 32 genes encoding thaumatin-like proteins in A. sativum cv. Ershuizao genome. The AsTLP genes were distributed among six chromosomes and four scaffolds and might have been evolutionary originated from segmental or tandem duplications. AsTLP7, 16, 20, 21, and 23 homologs were amplified from garlic cultivars resistant and susceptible to Fusarium infection. The promoters of AsTLP genes contained distinct sets of cis-acting elements associated with hormone and stress reactivity, suggesting differential transcriptional regulation of garlic TLPs in response to pathogens and abiotic stresses, which was consistent with specific expression patterns of AsTLP genes in garlic cultivars infected with F. proliferatum. The transcription of AsTLP7–9, and 21 genes in the roots was downregulated in FBR-susceptible and upregulated in FBR-resistant cultivars, suggesting their particular involvement in the sensitivity of garlic to fungal infection. Our results provide the foundation for further functional characterization of the AsTLP genes using a reverse genetics strategy, and may contribute in the breeding of A. sativum cultivars with increased resistance to Fusarium infections as well as various abiotic stresses.