Disease Diagnosis in Tea (Camellia sinensis (L.) Kuntze): Challenges and the Way Forward

Abstract

Derived from the Camellia sinensis (L.) Kuntze plant, tea is the most widely consumed natural beverage in the world. Tea is a perennial woody plant. Monoculturing tea on a large scale makes it susceptible to many perennial and seasonal diseases. The leaves, stems, and roots of tea plants are susceptible to fungal, bacterial, and viral pathogens. Tea is predominantly grown in Asian and African regions; hence, conventional methods including symptomology and signs, and microbiological and microscopic methods are mainly used in disease diagnosis and pathogen identification. Accurate and rapid identification of diseases and pathogens is imperative for the sustainability of tea plantations. Thus, the technological advancement in plant disease diagnosis also embraces the global tea industry. This chapter discusses new technological advances in tea disease diagnosis, focusing on molecular biological methods, whole genome sequencing, and remote sensing and image analysis methods. Further, it highlights the challenges in disease diagnosis as a perennial woody plant and pins down available opportunities that could be successfully adopted to overcome the issues.

1. Introduction

Tea (Camellia sinensis (L) O. Kuntze), belonging to the Family Theaceae, is a cash crop extensively cultivated in regions spanning Asia, Africa, and South America. Prominent tea-producing countries include India, China, Kenya, Sri Lanka, Vietnam, and Indonesia [1]. The global tea trade holds a substantial value of approximately USD 9.5 billion, serving as a crucial source of export earnings. Young succulent leaves are the harvestable part of the tea plant being used to manufacture the second most consumed nonalcoholic healthy beverage in the world. Tea production faces a significant challenge from microbial pathogens. These pathogens not only directly lead to crop losses but also shorten the plant’s lifespan, impair the quality of processed tea, and necessitate increased investments in disease management. Annual crop losses ranging from 10 to 30% have been documented due to microbial diseases in tea.

Diseases affecting tea plants are categorized as leaf or foliar diseases, stem diseases, and root diseases. A variety of pathogens such as fungi, bacteria, viruses, and algae are responsible for causing diseases in tea plants. However, fungal diseases pose substantial issues in major tea-producing nations, and about 350 fungal diseases are associated with tea plants [2]. Table 1 summarizes the most common diseases of the tea plant.

Diseases and causative organism/s	Symptoms	Signs
Leaf disease
Blister blight Exobasidium vexans	Initial symptom is the emergence of pale-yellow translucent spots. These spots transform into blisters leading to the formation of an indentation. Subsequently, the blisters undergo necrosis. Blister on tender stems produce distorted stems and shoot dieback.	Blisters are characterized by a white-colored hymenium on the swollen surface. Basidiospores are produced at the tips of basidia, located outside the leaf tissues.
Gray blight Pestalotiopsis like species	Appears as small, brown-colored spots that gradually enlarge and turn gray. The margins of the lesion retain their brown color, displaying concentric rings.	Black acervuli can be observed on the lesion.
Brown blight Colletotrichum spp.	Appear as water-soaked, yellowish-green lesions that subsequently enlarge and become deep brown necrotic lesions. The lesion margins may display a yellowish-green tint with concentric markings.	Black acervuli can be observed.
Red rust (Algal disease) Cephaleuros parasiticus, C. mycoidea)	Algae manifest as red spots (5–7 mm) on leaf, imparting a rusty appearance. Spots can become lichenized and turn in to white in color. The leaves take on a variegated appearance (yellow or white). Infected stems become more rigid and produce longitudinal cracks.	Rusty appearance is produced by the upright red filamentous structures carrying sporangia laden with spores.
Bacterial leaf and shoot blight Pseudomonas syringae pv. theae, Acidovorax avenae	Water-soaked spots appear first in young leaves. As disease progresses, spots merge together into large patches, killing the leaves. Symptoms in young shoots become visible after about 2 weeks of leaf infection.	Bacterial streaming under microscope.
Viral diseases Tea plant necrotic ring blotch virus (TPNRBV) and tea plant line pattern virus (TPLPV)	Leaf discoloration including albino and chlorina. Necrotic ring blotch can be seen. Premature defoliation of leaves may occur. Weaken growth.
Tea white scab disease Phyllosticta theaefolia, Elsinoe leucospila	Appear as small red spots with a grayish white or grayish brown center. As disease progresses, the spots become Bird’s eye shape, round necrotic spots. Under favorable conditions, spots cover the entire leaf and cause deformity and death.
Tea leaf spot Didymella segeticola var. camellia (Basionym: Phoma segeticola)	Appear as small pinhead-sized spots. Black-brown color spots can reach up to 1.2 mm in diameter. Leaf growth can be decelerated.
Alternaria longipes	Appear as round (4–6 mm in diameter), brown color spots. The center of the spot will turn into white in color.
Epicoccum nigrum	Spots first appear at the margins. Gradually expand and become irregularly shaped, dark brown color, necrotic lesion.
E. layuense	Small brown and necrotic spots on young and mature leaves. Spots later become large brown lesions.
E. sorghinum	Light or dark brown color, irregular spots appear at the margin. As diseases progress necrosis at the initial infection site can be seen.
Lasiodiplodia theobromae	Initially chlorotic spots appear that gradually become large irregular brown spots. Spots cover entire width of the leaf with bark brown or black center. Finally, dried leaves are dropped.
New leaf blight disease Muyocopron laterale	Reddish brown lesions at the leaf margin.	Sometimes black ascocarp can be seen on leaf surface.
Arthrinium arundinis	Appear as gray-brown dots. Dots expand to produce irregularly shaped dark brown color lesions (3–4 cm).
Alternaria alternata	Grayish brown patches first appear around tips and margins of young leaves that later expand toward the midrib. Leaf curl, death of leaves, and defoliation.
Stem diseases
Stem and branch canker Botryosphaeriaceae spp. (formally Macrophoma theicola)	At its onset, an oval, sunken, dark lesion appears that causes bark to decay and detach from the wood. Callusing around the lesion’s edges forms thickened bark areas known as cankers which may gradually expand and encircle the stem. Under severe infection, leaf yellowing, necrosis, dieback, and eventual leaf shedding can be seen.	Perithecia, varying in color from ash to black, are infrequently sighted as the disease progresses to more advanced stages such as wood decay.
Collar canker and die back Phomopsis theae, Fusarium solani species complex	Cracks and bark peeling occur in the collar region, either at the soil level or just below it. Heavy flowering and fruiting are commonly followed by canopy thinning. Progressive dieback of branches results in bare bushes.	Under damp and moist conditions, reddish, tiny superficial perithecia are visible in bushes infected by F. solani species complex.
Root disease
Red root disease Poria hypolateritia (current name – Ceriporiopsis hypolateritius), Ganoderma philippine	Infected roots exhibit a mottled appearance. Over time, both the root surface and inner tissues turns through colors of white, red, and eventually brick red. With time, the infected roots disintegrate into a soft pulp. Leaf yellowing, wilting and dieback.	A fine layer of white mycelium and white mycelial cords are noticeable between the wood and the bark and on the root surface, respectively. As time progresses, these white mycelial cords turn in to red and fuse together, giving rise to a network that spreads across the root.
Brown root disease Phellinus noxius	The roots are encrusted with a compact mass of soil and small stones that are difficult to remove by washing or rubbing. As disease progress, the inner tissues of the roots decay, resulting in the formation of a honeycomb-like structure. Leaf yellowing, wilting and dieback.	White to brown woolly masses of mycelia are found attached to the encrusted mass of earth. These mycelial masses gradually turn in to dark brown in color.
Black root disease Rosellinia arcuate, R. bunoides	Gray to black coloration is observed on the root surface, extending to the collar region and above in some cases. The stem’s bark near the soil surface may be found dead, spanning a length of 7.5–10 cm. Leaf yellowing, wilting and dieback.	White to black strands of mycelia firmly adhere to the roots, forming a loose cobweb-like mass. Numerous small, white, star-shaped mycelial structures develop on the wood’s surface beneath the bark.
Armillaria root rot Armillaria spp.	Premature flowering and somewhat stunted growth are evident. Longitudinal cracks are typically present above the soil level on the collar, as well as on the taproot and lateral roots. Leaf yellowing, wilting, and dieback.	Creamy white, fan-shaped mycelia can be seen under the bark. Additionally, dark brown to black rhizomorphs with a white interior develop beneath the bark. Clusters of 5–21 basidiomata can be seen.
Charcoal root disease Ustulina zonata U. deusta	Black, irregular double lines are visible in the wood. This is often accompanied by leaf yellowing, wilting, and sudden death of the bushes.	White color, large, fan-shaped patches of mycelium are seen on wood surface upon the removal of bark.

Table 1.

Use of symptoms and signs in the diagnosis of tea diseases.

1.1 Leaf diseases

Blister blight (Exobasidium vexans), brown blight or anthracnose (Colletotrichum species complex), gray blight (Pestalotiopsis-like species), and bacterial blight (Pseudomonas syringae pv. theae and Acidovorax avenae) are the major foliar diseases of tea. The diseases are problematic in all the tea-growing countries in Asia. Gray blight is problematic in African tea-growing countries as well. The crop loss recorded from blister blight and gray blight ranges from 20 to 50% and 10 to 20%, respectively. The extent of crop loss caused by the other diseases is considerable, even though exact figures are unavailable. In addition to crop loss, leaf diseases impact the quality of the final product by altering the biochemical constitution of the leaves. Foliar diseases become problematic generally under cool and humid weather conditions.

1.2 Stem diseases

Cankers are widely reported stem diseases in tea, prevalent in almost all tea-growing countries. The pathogen/s attack twigs and stems, kill or griddle branches, and ultimately kill the entire bush. Cankers on the stems and or branches are referred to as stem and branch cankers. Cankers that originate from the collar region near the soil surface are termed collar cankers (Figure 1I–J). Botryosphaeriaceae fungi Lasiodiplodia theobromae, Botryosphaeria dothidea, and B. mamanae are reported to cause stem and branch cankers. Phomopsis theae (Diaporthe) and Fusarium solani species complex have been identified as pathogens that cause collar cankers. Cankers are persistent and highly influenced by soil conditions, dry weather, plant nutritional status, and agronomic practices. Hypoxylon wood rot, caused by Nemania diffusa, and thorny stem blight, caused by Tunstallia aculeata (formally known as Aglaospora aculeata) are two other less common stem diseases.

Figure 1.

Symptomology of tea diseases. (A) Blister blight, (B) brown blight, (C) gray blight, (D) red rust disease, (E) Phyllosticta leaf blight, (F) red root disease, (G) black root disease, (H) brown root disease, (I) stem and branch canker, and (J) collar canker.

1.3 Root diseases

Root rots are widespread among tea-growing countries. Root infections result in the sudden death of branches or entire bushes. Some root infections lead to the gradual death of bushes following fruiting, flowering, and defoliation. Armillaria spp. Poria hypolateritia (Ceriporiopsis hypolateritia), Rosellinia arcuata (Dematophora arcuata), Phellinus noxius (Pyrrhoderma noxium), and Ustulina zonata (Kretzschmaria zonata) are major root pathogens in tea. The symptoms of infection expressed on the root by each of the pathogens are unique (Figure 1F–H).

Accurate disease diagnosis, identification of the pathogen, and disease quantification are crucial steps at the forefront of effective disease management. Early diagnosis assumes paramount importance as it enables timely intervention and targeted treatment strategies. Furthermore, accurate estimation of disease incidence and severity, along with a comprehensive study of the detrimental impacts of diseases on both harvest quality and quantity, has gained significant importance in various facets of crop cultivation. This chapter summarizes the novel and advanced approaches now in use in disease diagnosis and pathogen identification and quantification in tea and sheds light on the general approaches in this area.

2. Techniques used in tea disease diagnosis

2.1 Symptomology and signs

Symptomology, a fundamental and oldest tool in plant disease diagnosis, involves the observation and interpretation of visible alterations in plant appearance caused by pathogens. By closely examining leaf, stem, and root discoloration, lesions, deformities, wilting, the presence of pathogen structures, and other abnormalities, such as cankers, rots, etc., pathologists have long been able to diagnose diseases efficiently.

The symptomology and signs of various widespread tea diseases are well-documented (Table 1 and Figure 1). Yet, tea disease diagnosis is frequently complicated by overlapping symptoms, leading to confusion. For example, gray, brown, and black lesions on the leaf surface are defined symptoms of gray blight, brown blight, and black blight diseases, respectively. Nevertheless, the subjective nature of color recognition can introduce confusion in disease diagnosis, particularly considering the absence of distinct demarcations among these three colors. Further challenges arise with newly reported leaf spot diseases that often exhibit similar symptoms [3, 4]. Diagnosing stem and root diseases in tea is exceptionally challenging, especially during the early stages of infection. Symptomology and signs of these diseases are often concealed beneath the densely covered tea canopy and the soil, respectively. Even when diseases progress, foliar symptoms tend to resemble one another, including leaf yellowing, necrosis, and branch and bush dieback. Consequently, meticulous observation of symptomology and signs on stems and roots often requires the uprooting or removal of branches, highlighting the difficulty of diagnosis. This process involves labor-intensive field operations. On the other hand, the distinct symptoms that are being used to diagnose root diseases appear at later stages of disease progression. Hence, early diagnosis using this method is extremely challenging.

2.2 Microbiological and microscopic methods

Microbiological and microscopic methods commonly involve the collection of samples from infected plant tissues, followed by culturing them on specialized microbiological media to isolate and identify the target pathogen(s). This process relies on observing the culture morphology and characteristics of asexual and sexual structures under a microscope for accurate identification. Additionally, microscopic methods can be employed to study infected plant tissues, focusing on identifying tissue deformities and the presence of pathogen structures within the tissues.

Culturing and microscopy for correct diagnosis hold significant importance in tea pathology. There are scenarios where a single disease or closely resembling symptoms can be caused by more than one species of pathogens. For instance, multiple species of Colletotrichum and Pestalotiopsis have been identified as pathogens of brown blight and gray blight diseases. Leaf spot disease can be caused by Didymella segeticola, Alternaria longipes, and different species of Epicoccum [5, 6, 7]. Several species within the Botryosphaeriaceae family have been identified as pathogens causing stem and branch cankers in Sri Lanka [8]. Collar canker disease can be caused either by P. theae or F. solani species complex [9]. Poria hypolateritia (currently classified as Ceriporiopsis hypolateritius) and Ganoderma philippine have been documented as the causes of red root disease. Hence, in the absence of well-developed novel technologies, accurate diagnosis of most tea diseases relies heavily on microbiological and microscopic methods.

Since the majority of fungal tea pathogens are necrotrophs, they can be cultivated on commonly used microbiological media like potato dextrose agar (PDA). These pathogens thrive under standard environmental conditions, displaying specific cultural morphologies that are easily recognizable to skilled pathologists. For example, P. hypolateritia, P. noxius, and R. arcuata yield white color colonies on PDA. Over time, these colonies transform into milky white, yellow to brown, and black colors, respectively. Such distinctive color changes provide valuable cues for identification up to the genus level (Figure 2).

Figure 2.

Culture morphology (upper surface of 10 days old culture) of tea root pathogens (A) Poria hypolateritia (red root disease), (B) Rosellinia arcuata (black root disease), and (C) Phellinus noxius (brown root disease).

Light microscopy has been used for disease diagnosis in tea since the blister blight pathogen was first identified in 1898. Fluorescence microscopy and electron microscopy have been used to study the infection process of E. vexans [10] and to study the difference between E. vexans infected and healthy tea leaves [11]. Fluorescence microscopy using calcofluor white and wheat germ agglutinin stains and scanning electron microscopy of E. vexans inoculated tea leaves has been successfully used as a tool to study and diagnose the germination and infection stages of the pathogen [10].

2.3 Molecular biological methods

Molecular biological methods in plant disease diagnosis involve the analysis of genetic material, such as DNA or RNA, to identify plant pathogens and eventually diagnose diseases. These techniques have gained popularity in plant pathology due to their accuracy and specificity. Common molecular biological methods used for plant disease diagnosis include polymerase chain reaction (PCR) and its variants, loop-mediated isothermal amplification (LAMP), nucleic acid hybridization, nucleic acid amplicon sequencing and phylogenetics, CRISPR-Cas-based methods, metagenomics, etc.

2.3.1 Loop-mediated isothermal amplification (LAMP)

Loop-mediated isothermal amplification is a rapid disease diagnosis tool becoming increasingly popular in plant pathology due to its high sensitivity, cost-effectiveness, and ability to visually verify positive results. A LAMP reaction consists of three steps, namely the initial step, the cycling amplification step, and the elongation step. In this method, six unique sequences present in the targeted nucleic acid are identified by a set of two internal primers, one backward loop primer, and a set of two outer primers. In tea, LAMP methodology has been developed for the detection of the blister blight pathogen E. vexans [12] and on-site diagnosis of the anthracnose pathogen C. siamense [13].

2.3.2 Polymerase chain reaction

Polymerase chain reaction has revolutionized plant disease diagnosis and pathogen detection by offering a highly sensitive and specific tool for identifying the presence of pathogens in plants. PCR operates by amplifying specific segments of the pathogen’s nucleic acid through a series of temperature cycles, including denaturation, annealing, extension, and final extension. By utilizing primers that are specific to the target pathogen’s genetic material, PCR allows for the identification of pathogens. Conventional end-point PCR, quantitative PCR (qPCR), multiplex PCR, and nested PCR are some variations of PCR methods that are widely used in plant pathology. PCR products are subjected to sequencing and compared with the existing data in the NCBI GenBank. This comparison process allows for the development of species-specific primers as needed, ensuring a more precise and tailored approach for pathogen identification.

The sequence information for most tea pathogens is limited. Hence, universal primers are used in most studies investigating the molecular identification and diversity of tea pathogens. Using the 28SrRNA sequence of E. vexans available in the NCBI GenBank, Chaliha et al. [10] developed a species-specific primer pair for the selective amplification of E. vexans (EVLSUF-EVLSUR). However, the combination of universal forward primer ITS1F and species-specific reverse primer EVLSUR allows for the identification of multiple strains, resulting in a larger PCR product.

An RT-qPCR method has been developed to quantify the C. camelliae (tea brown blight or tea anthracnose disease) fungal biomass in tea leaves. This method uses species-specific primers (Table 2) and tea plant-specific primers (forward primer: 5′-GACTCCGCTGGCACCTTAT-3′; reverse primer: 5′-GCCCTTCCGTCAATTCCT-3′) for normalization [14]. Compared to the traditional lesion area measurement approach, the proposed RT-qPCR method is highly sensitive and rapid. Further, this method allows for the differentiation of small differences in disease severity levels that may be undetectable by lesion size alone.

Pathogen [Reference]	Method	Name	Primer sequences
Exobasidium vexans; [10, 12]	LAMP	Inner primers (EV_FIP, EV_BIP)	F-TACGGGGCTCTCACCCTCTATG + TGTAGAGGATGCTTTTGGCG; R-GACCACCGAGCCTCTGTAAAGC + AGAAGACGTACACCTCCCAT
		Loop primers (EV_LF, EV_LB)	F-TTCCAGGGAACTCGGAAGGCA; R-TCCTTCGACGAGTCGAGTAGTTTGG
		Outer primers (EV_F3, EV_B3)	F-TCTCTGGCCGGTATTTAGCT; R-TGAAATCTGGCCGCAAGG
	End-point PCR	EVLSUF-EVLSUR	F-GCCGGATGAGCTCAAATTTG; R-CCAAAACTGATGTTGGCCTG
		ITS1F-EVLSUR	F-CTTGGTCATTTAGAGGAAGTAA; R-CCAAAACTGATGTTGGCCTG
Colletotrichum siamense; [13]	LAMP	Inner primers (FIP, BIP)	F-GAAAAAGCACCGTCGTAGCCA+ TGAAGGATACCGACTCCGAA R-ATATGTTGACTCTGAATAGG+ ATGACGTGACGCAACTCG
		Loop primers (F-loop, B-loop)	F-AGCCTCGCGAATCTCCTC; R-TCTTTGACCGCGACAACAAC
		Outer primers (F3, B3)	F-ACTATGATGGCCCGCAAGA; R-GCTTCTCACCAATGGAGGTC
	End-point PCR	P-Cs	F-TGCTTTTTCGCTGCGATAT; R-CCATGACGCATACCAATCAA
C. camelliae; [14]	RT − qPCR	S572/S573	F-CCCGCATCTGGTAGACAAGA; R-TGATAGCATGTGTCCCTCCG
		S576/S577	F-AAAGGTAGTGGCGGACCCT; R-CCCAGTGCGAGACGTAAAGT
Tea Plant Necrotic Ring Blotch Virus; [15]	RT − qPCR	qTPNRBV1	F-TGTGCGACGGACTCATTCTC; R-TCAAGCATGCGAGTTTCCCT
		qTPNRBV2	F-CCCGTCATCGATACGCTCAA; R-TACCTGGGGGAAGGTCTGAA
		qTPNRBV3	F-GATCGTCCTTCCCGATCTGC; R-GTGCAGGCGTTCGCTTATTG
		qTPNRBV4	F-TTGGTTCGCTGTCCACCATT; R-GAGCTGAAGCTGACGCATGA
Tea plant necrotic ring blotch virus; [16]	RT-PCR	TPNRBV1	F-GCCCTGACAACGCAAAAGAACTGATG; R-GTGACGGGATATTTTTGGACGACTGT
		TPNRBV2	F-GGGCCGGGGTGTGGAAAAACTT; R-TTCTTATCATCCCGGCAAAACACA
		TPNRBV3	F-TTCGCCACTCACAAAGACAACAAACT; R-GTAGCGGAGCGGAAAGAAAAGACT
		TPNRBV4	F-TCAGTGGCGCGATTATCAGAAGGTA; R-CGCGCAAGAAGTCGGTCAAAAC
[17]	RT-PCR	TPNRBV	F-CCTTATGTCGACAGTTGCTAC; R-CTAAGTCATCCATATGTGTGG
Tea plant line pattern virus; [16]	RT-PCR	TPLPV1	F-AAGGTGGCGAGGTCAGTTTCAGTTCA; R-TCCCCATAGGTTCATCTTGTAGCAGTCG
		TPLPV2	F-CCTATGGAGCTCTATGACGCAAATGAT; R-TCGACAGAGAAGTGATGGGGAAATAC
		TPLPV3	F-TCAGATGCGACCAATGGAACTCAACT; R-CCCGCGCTAATCCTCTCAAGACTAACTA

Table 2.

Species-specific primers for polymerase chain reaction (PCR) and loop-mediated isothermal amplification (LAMP)-based diagnosis of fungal and viral disease in tea.

F, forward primer, R, reverse primer.

Species-specific internal transcribed spacer (ITS) markers have been developed for stem and branch canker pathogens B. dothidea and L. theobromae, allowing detection through conventional PCR methods. This method is useful for identifying the latent infection of the pathogen in planting material and targeting species-specific fungicide treatment in infected tea plants [8].

Viral diseases are uncommon in tea. Recently, Tea Plant Necrotic Ring Blotch Virus (TRNBV) and Tea Plant Line Pattern Virus (TRLPV) have been identified in China. The genome of TRNRBV has four positive-sense single-stranded RNA, and TRLPV has three segments of single-strand RNA [16, 18]. Conventional and a SYBR Green RT-qPCR method have been developed for the detection of four different segments of TPNRBV RNA [15, 16]. Importantly, the RT-qPCR method offers a significant advantage: it is 100-fold more sensitive than conventional detection methods. Table 2 provides details of specific primers for PCR and LAMP-based diagnosis of diseases in tea.

2.3.3 Metagenomics

Metagenomic sequencing is a cutting-edge technique in which the genetic material of entire communities of microorganisms in a particular sample is evaluated. This method delves into the hidden and complex world of microorganisms by directly sequencing and analyzing DNA or RNA extracted from a particular sample. In plant disease diagnosis, metagenomics provides an unbiased, comprehensive approach to identifying pathogens associated with the disease symptoms. This technique has the added advantage of identifying unculturable, unexpected emerging pathogens often missed by standard approaches. The first application of the metagenomics approach to pathogen identification in tea was performed in China. In this study, researchers identified seven putative viruses including two novel species, TPNRBV, and TPLPV from tea leaves exhibiting albino and chlorina symptoms [16].

2.3.4 Nucleic acid amplicon sequencing and phylogenetics

Nucleic acid sequencing and phylogenetic analysis have revolutionized plant disease diagnosis. These tools allow for accurate pathogen identification, disease origin tracking, and a deeper understanding of the evolutionary relationships between different pathogen strains. By sequencing specific genetic markers (e.g., 16S ribosomal RNA (16S rRNA) for bacteria and internal transcribed spacer (ITS) within ribosomal RNA for fungi) and comparing them against a reference database, pathogen/s can be precisely identified. While 16S rRNA and ITS are widely used, they often lack the resolution needed to distinguish closely related strains. In these cases, sequencing additional marker genes and follow-up phylogenetic analysis significantly increases taxonomic accuracy.

DNA sequencing and phylogenetic analysis have significantly advanced tea pathology, revealing unknown pathogens and correcting misidentifications. For instance, increasing leaf blight reports from India and China show symptom similarity, leading to misdiagnosis. DNA sequencing revealed at least three distinct pathogens, Muyocopron laterale [3], Alternaria alternata [19], and Arthrinium arundinis [20]. Similarly, diverse pathogens were identified for leaf spot diseases across China (Table 1) [4, 5, 21, 22, 23].

Macrophoma theicola was long thought to cause stem and branch canker. However, recent studies utilizing DNA sequencing and phylogenetic analysis have refined this identification to three species within Botryosphaeriaceae family [8]. Controversy surrounds the identity of the pathogen causing tea white scab disease, with some suggesting Phyllosticta theaefolia or Elsinoe leucospila. However, a comprehensive approach using DNA sequencing, morphological characterization, and pathogenicity tests identified E. leucospila as the prominent pathogen [24].

Multigene phylogenetic analysis has revealed that gray blight disease in tea is caused by a diverse group of species belonging to at least three genera: Pseudopestalotiopsis, Neopestalotiopsis, and Pestalotiopsis [25, 26, 27, 28, 29, 30, 31, 32]. According to the earliest known report by Massee from Sri Lanka, brown blight disease in tea is caused by C. camelliae. However, recent multi-locus phylogenetic approaches have revealed at least 17 Colletotrichum species can cause the same disease with varying degrees of severity [33, 34, 35, 36, 37, 38].

2.3.5 Whole genome sequencing

Whole genome sequencing (WGS) provides a comprehensive analysis of a pathogen’s genetic blueprint, providing in-depth insights into its biology, virulence factors, and evolution. Unlike PCR-based methods and amplicon sequencing, whole genome sequencing can be conducted without the need for specific primers. In tea, genome sequencing of only very few pathogens is available. However, there is no evidence to show its application in pathogen identification yet. Kong et al. [38] sequenced the whole genome (67.74 Mb) of tea brown blight pathogen C. camelliae using the PacBio RSII sequencing platform and identified the pathogenicity-related gene clusters. The genome of tea leaf spot pathogen D. segeticola (33.4 Mb) has been sequenced, and its importance in studying pathogenicity determinants, pathogen evolution, and disease control strategies has been highlighted [15]. The whole genome of Ps. theae (50.41 Mb) has been sequenced and deposited at GenBank under the accession number JACWGD000000000 [39]. Whole genome sequences of two different strains of tea plant virus TPNRBV (Chinese isolate and Japanese isolate) are available [16, 18]. This information proved valuable to the design of species-specific primers for PCR-based diagnosis of viral diseases in tea.

2.4 Remote sensing and image analysis methods

Remote sensing and image analysis have emerged as powerful tools in the field of plant disease diagnosis, revolutionizing the way we detect, monitor, and manage diseases in agricultural and natural ecosystems. Remote sensing involves the use of specialized sensors attached to satellites, drones, or ground-based instruments, to capture data from a distance. These sensors collect a wide range of data, including optical, thermal, and hyperspectral images. Infrared thermal imaging, red, green, and blue (RGB) imaging, multispectral imaging, hyperspectral imaging, chlorophyll fluorescence imaging, 3D sensors, etc., are some examples of remote sensing and imaging technologies widely used in plant disease diagnosis.

2.4.1 Multispectral and hyperspectral imaging

Spectral imaging provides a marker for a plant’s health by using single or multiple wavelengths in the electromagnetic spectrum. Healthy plants with their higher chlorophyll content absorb more red light and therefore reflect a higher proportion of near-infrared (NIR) than diseased plants. A diseased plant does the opposite. The information is combined in the form of different vegetation indices to estimate the plant’s fitness or performance.

Hyperspectral imaging technology has been employed to identify anthracnose disease in tea [40, 41]. Compared to healthy leaves, diseased leaves exhibited reduced spectral reflectance near 550 nm and 770 nm, with an increase observed around 680 nm. The accuracy of the method was 94.12–94.28% [40].

Hyperspectral signatures in the wavelength range of 700 to 1000 nm (NIR region) differentiated blister blight-infected leaves from healthy ones [42]. Blister blight disease progresses through the translucent spot, mature blister, and necrotic stages. The reflectance decreased by 2.4, 4.5, and 30.3% as the disease progressed from the translucent spot to the necrotic stage, respectively. Spectral wavelengths at 560, 640, and 780 nm differentiated anthracnose, brown leaf spot disease, and tea white star diseased leaves from healthy leaves [43]. A hyperspectral image classification model CARS-LSTM (The competitive adaptive reweighted sampling-Long short-term memory) achieved 95% accuracy in identifying tea coal disease (Neocapnodium theae Hara), outperforming the RGB model, WT-ResNet18 [44].

2.4.2 RGB imaging

RGB imaging in plant disease diagnosis involves capturing digital images of plants using standard cameras, including smartphones and specialized imaging devices, and follow-up analysis and comparisons of its three features: color, texture, and shape. Artificial Intelligence (AI) and Machine Learning (ML) play pivotal roles in image analysis, revolutionizing how we interpret and extract meaningful information from visual data.

In the recent past, there has been a noticeable rapid increase in the use of RGB image analysis for tea disease diagnosis. The targets are foliar diseases, including blister blight, gray blight, brown blight/tea anthracnose, tea coal disease, algal leaf disease, leaf spot diseases, etc. Importantly, multiple diseases together with pest-infested leaves, were utilized to improve the differentiation potential and classification accuracy. Further, the above approach has laid the foundation for diagnosing multiple diseases and pest attacks on the same tea leaf. Among ML algorithms, Convolutional Neural Networks (CNNs) stand out as a popular choice, primarily attributed to their exceptional capabilities in image classification (Table 3).

Diseases	Classification algorithm/architecture	Average accuracy	Ref.
Gray blight	Deep CNN based on MobileNet and VGG19 Net	98.99%	[45]
Gray and brown blights, algal, and red spots, Helopeltis pest attack	Deep CNN	96.56%	[46]
Red rust, red spider, thrips, and Helopeltis pest attack	Multi-class support vector machines	83%	[47]
Red, white, and algal leaf spots, bird’s-eye spots, gray and brown blights	Deep CNN model (LeafNet)	90.16%	[48]
Algal leaf disease	Ensemble neural network	91%	[49]
Brown blight and algal spots	Support vector machine classifier	90%	[50]
Leaf blight diseases	Faster region-based convolutional neural networks (Faster R-CNN), and VGG16	84%	[51]
Tea red scab, tea red leaf spot, and tea leaf blight	Conditional deep convolutional generative adversarial networks (C-DCGAN), VGG16	90%	[52]
Tea red leaf spot and tea red scab	Gradient Boosting Decision Tree (GBDT)	95%	[53]
Leaf and bud blight and red scab	Custom DCNN	92.5%	[54]
Leaf lesions	Faster R-CNN and cascade R-CNN (CRCNN)	74–76.6%	[55]
Blister blight	Deep hashing with integrated autoencoders (DHIA) consisting of basic model (BM) and autoencoders (AE)	>97%	[56]
Algal leaf spot, gray and brown blight, white spot, red scab, and bud blight	CNN	94.45%	[57]
Algal leaf spot, anthracnose, and bird’s-eye spot	CNN	95%	[58]
Algal leaf spot, leaf and bud blight, white scab	AX-RetinaNet	93.83%	[59]
Leaf blights	Lightweight and efficient LC3Net model with integration of a channel attention module (CAM)	92.29%	[60]

Table 3.

RGB imaging techniques and machine learning algorithms are used for artificial intelligence-based diagnosis of tea diseases.

Disease monitoring is also achieved with imaging and ML techniques. Analysis of RGB images using YOLOv8 and Residual Network 50 (Resnet50) CNN models have been effective for diagnosing and assessing the severity of blister blight disease [61]. Edwin Raj et al. [62] used RGB images to develop a methodology to spatially assess the percentage of disease with 95% accuracy. Furthermore, the method facilitates the mapping of disease occurrences at the estate level using smartphone GPS data and Geographic Information System (GIS).

Dahanayake et al. [63] developed an Android app for diagnosing tea leaf diseases using smartphone images. This app with Artificial Neural Networks (ANNs) allows farmers to diagnose diseases on-site, saving time and money. A study developed a CNN model and deployed it to the Platform-as-a-Service cloud for classifying tea leaf disease using smartphone images. Hence, the digital images captured using smartphones can be uploaded to predict the disease automatically [64].

3. Challenges in tea disease diagnosis

Tea is grown as an intensive monoculture crop. This system is vulnerable to the rapid spread of pests and diseases, making the problem unmanageable within a short time. Early detection and accurate identification of diseases and their causal agents are paramount for efficient and economically viable disease management. With the advancement of science, a combination of traditional and modern techniques is used in the diagnosis of tea diseases. However, there are several challenges to be faced. This section analyses the major challenges faced in tea disease diagnosis.

3.1 Disease symptoms with a wide range of characteristics

Some diseases may produce easily recognizable symptoms such as discoloration, wilting, or visible lesions, while others may exhibit subtle or less distinct symptoms that are easily overlooked. For instance, blister blight disease develops through different stages over 2 to 4 weeks. Though the leaf and stem blisters are easily identified and captured by various techniques, the first visible tiny to small translucent spots generally go unnoticed and are difficult to sense by imaging techniques. This makes early detection of the disease challenging even with modern imaging techniques.

3.2 Similar symptoms produced by different diseases/factors

Leaf blights caused by different pathogens show symptoms that overlap (Section 2.1). For example, brown blight and gray blight share common features in their symptom expression. Delimitation of these two diseases only by symptomology is always impossible. Under such conditions, culturing techniques or DNA-based techniques help with accurate identification. Leaf spot diseases also share common phenotypical expressions. Nutrient deficiency symptoms resemble leaf blights and leaf spots.

Sudden death of a branch or whole plant with attached leaves occurs when tea plants are infected with root pathogens. Though root symptoms may be different, the above-ground symptoms are common for most of the root diseases. Death of plants due to drought, white grub infestation, and root diseases also share similar features and always demand careful diagnosis for differentiation.

3.3 Simultaneous occurrence of more than one disease

Under certain environmental conditions, co-infections of fungal pathogens are common. Brown blight infection generally follows blister blight disease. The occurrence of brown blight and gray blights on the same leaf or plant is a common feature. The application of imaging techniques or DNA-based techniques is difficult under such conditions.

3.4 Tea canopy architecture and leaf orientation

Naturally, the tea plant grows as a tree in the wild. Periodic pruning and frequent harvesting maintain the plant as a bush. Close planting (12,500 plants per hectare in Sri Lanka) helps to form a continuous harvesting table. There are several generations (different age groups) of tea shoots at a time on the tea canopy. The overlapping shoot arrangement and leaf orientation always challenge image-based techniques in disease diagnosis.

Similarly, stem diseases such as cankers, wood rots, and stem blights cannot be diagnosed early with imaging techniques or monitoring at a distance. Careful observation of stems and branches is necessary to identify such conditions, as they may only become visible during pruning operations.

3.5 Asymptomatic infections

Several species of Botryosphaeriaceae, Pestolotiopsis, Phomopsis (Diaphorthe), etc., occur as endophytes or latent infections in tea plants where the pathogen(s) are present in the plant but do not cause visible symptoms. This can lead to undetected infections and disease spread.

3.6 Environmental factors, plant nutrient status, and plant factors

Environmental conditions, such as weather, humidity, soil quality, and plant nutrient status, can influence the development and appearance of disease symptoms. Different tea plant varieties respond to certain diseases differently, which can complicate diagnosis because the same pathogen may have different effects on different varieties. These variabilities can make it challenging to establish a definitive diagnosis protocol.

3.7 Limited diagnostic tools and expertise

The global tea industry slowly and gradually grabs new and modern technologies as tea is mostly produced in relatively low-income countries [65]. There is limited access to advanced diagnostic tools, such as imaging and image analysis, DNA-based tests, or specialized equipment and trained personnel, which are essential for accurate disease diagnosis in most of the tea-producing countries.

3.8 Climate change

Tea is produced in narrowly defined agroecological conditions as a rain-fed monocrop. The changes in the rainfall patterns, rising temperature, and increasing events of floods and drought events would lead to the emergence of new pathogens. The possible emergence of new diseases and changes in existing pathogen populations would make disease diagnosis challenging, demanding more sophisticated methods and techniques.

4. Way forward

It is forecasted that the global tea industry will grow with a compound annual growth rate (CAGR) of 6.7 percent from 2022 to 2031. Minimizing crop loss due to pests and diseases while ensuring pesticide-free quality products is a challenge to the growing tea industry under current climate change. There are several opportunities for improving tea disease diagnosis, which can lead to more effective disease management and increased crop yield in the tea industry. These opportunities encompass technological advancements, research, and collaboration.

As many tea-growing countries share common disease problems, global collaboration in disease diagnosis and studies on pathogens would help to bring appropriate technology rapidly at a shared cost. Once such technology is developed with coordinated effort, refinement at local conditions can be carried out. This collaborative approach could minimize duplication of work among tea-growing countries on the same disease problem. The traditional method of disease diagnosis with culturing and disease symptomology has been supported by DNA-based methods in the recent past for many tea diseases. The available information can be used by tea pathologists with modification if required.

The establishment of disease surveillance networks that involve collaboration among tea growers, researchers, and government agencies, along with timely sharing of disease data, can facilitate rapid response and containment. Training and educational programs for tea growers and plantation workers to enhance their knowledge of disease symptoms, prevention, and early detection techniques could help manage resource limitations.

Whole genome sequences are available only for brown blight pathogen C. camelliae, tea leaf spot pathogen D. segeticola, P. theae, and two different strains of tea plant virus TPNRBV despite more than 300 fungal pathogens being reported in tea. With global collaboration and wise utilization of resources, it is imperative to generate whole genome sequences of major tea pathogens. This information would help develop and implement advanced diagnostic tools, as well as high-throughput assays. These technologies can provide rapid and highly accurate identification of tea pathogens.

Although imaging and remote sensing technology are developing rapidly, most of the works are from academic research limiting agricultural applications. Hyperspectral imaging may be promising due to its high specificity and rapid data analysis. However, the use of drones or unmanned aerial vehicles for collecting data would remain challenging in situations where tea is grown under shade trees, particularly in India and Sri Lanka. To get the maximum potential of machine learning techniques, a large, verified dataset of images of diseased and healthy plants is essential. The establishment of quality data sets may overcome the problems faced by data scientists. Smartphone and mobile phone-based diagnostic tools will help share knowledge and improve the diagnostic capabilities of even the tea growers.

There are unexplored horizons in tea disease diagnostic methods. Nanotechnology-based pathogen detection, biosensors based on highly selective bio-recognition elements, and CRISPR-Cas-based detection systems have so far not experimented with tea pathogens. Although the application of ELISA for tea disease detection has not been reported much, the fabrication of test strips based on ELISA will be an innovative strategy for the detection of tea viruses in the future due to its ease of handling. Lateral flow assays would favor in-field quick diagnosis. It would be advantageous to develop multiple pathogen detection systems rather than the detection of a single pathogen.

5. Conclusions

An early and precise diagnosis of diseases is crucial, enabling prompt intervention and precise disease management plans. The global tea industry is gradually embracing modern methodologies rather than relying solely on symptomatic and culture-based methods. DNA and RNA-based diagnosis, genome sequencing, remote sensing, machine learning, and image processing technology to achieve automatic detection are widely used in tea. The adoption of current techniques and their field-level applications are, however, influenced by several factors. The future needs are discussed in detail.

Conflict of interest

The authors declare no conflict of interest.