Evidence for Reconsidering the Taxonomic Status of Closely Related Oligonychus Species in punicae Complex (Acari: Prostigmata: Tetranychidae)

Simple Summary The present study comprehensively addressed a taxonomic problem in the Oligonychus punicae species complex. Based on the morphological and molecular data, two spider mite species, namely Oligonychus mangiferus (Rahman and Sapra) and O. vitis (Zaher and Shehata) are synonymized with O. punicae (Hirst). Moreover, DNA-based analysis showed that there are GenBank COI and ITS2 sequences of Oligonychus that need to be revisited to confirm the identity of their voucher specimens. The importance of an integrative taxonomic approach is discussed for the accurate identification of closely related Oligonychus species. Abstract To elucidate the taxonomic problems in species delineation within the Oligonychus punicae complex (O. punicae, O. mangiferus, and O. vitis) (Acari: Prostigmata: Tetranychidae), we performed morphological and molecular investigations on mite samples, collected from different hosts/countries. Thirty-nine samples of punicae complex, collected from Egypt, Pakistan, and Saudi Arabia (SA), did not show any considerable morphological differences in females and males. All 39 samples of the punicae complex resembled the original description of O. punicae, while the claimed Mexican O. punicae was distinctively different based on male aedeagus. Molecularly, the low nucleotide diversity ranged from 0% to 2.1% (ITS2-rDNA) and 0% to 1% (COI-mtDNA), and was observed among various DNA sequences of the punicae complex from Egypt, India, Israel, Pakistan, and SA, confirming their identity as one species. The high genetic divergence ranged from 17.2% to 18.8% (ITS2) and 9.2% to 10.2% (COI), observed between the claimed Mexican O. punicae and all other sequences of the punicae complex, indicating that the Mexican sample do not belong to O. punicae. Basing our findings on both morphological and molecular data, we can conclude that O. mangiferus and O. vitis are synonymized with O. punicae. Additionally, this study reveals that the claimed Mexican O. punicae needs to be re-identified.


Introduction
Oligonychus Berlese is the largest genus of spider mites in the family Tetranychidae Donnadieu (Acari: Prostigmata), with 212 species that have been reported so far [1,2]. Differentiation of Oligonychus species is difficult and usually depends on the morphology of the male aedeagus [3,4]. However, both adult sexes are compulsory for the identification of the Oligonychus species. Indeed, exact species identification is usually challenging because of the minute differences in male aedeagus and the limited potential diagnostic characteristics available in females of several Oligonychus species [3][4][5][6][7]. Moreover, the key differential character of the male aedeagus becomes unreliable with improperly mounted specimens (not in a lateral position) or brief morphological descriptions without illustrations of the Forty spider mite samples from the punicae complex were collected from seven host plants in different seasons from various localities within four countries ( Figure 1; Table S1). Most of the samples were collected during the period from 2017 to 2022, except for a sample (voucher no.: 105; Table S1) that had been previously collected in 2012 and was identified as O. punicae [36] from Saudi Arabia (SA). A total of 26 samples (voucher no.: 1,5,6,13,14,21,26,27,37,43,44,56,59,62,63,64,70,76,77,78,80,104, 105, 134, 137 and 147; Table S1) were collected from various localities within eight SA provinces of Asir, Jouf, Jizan, Makkah, Madinah, Qassim, Riyadh, and Tabuk ( Figure 2 [37] were brought from Egypt and Mexico, respectively (Table S1). Some of the collected samples (voucher no.: 6,27,70,72,74,75,78,104,134,137,147,184, and 185; Tables S1 and S2) only contained female specimens. The collection details (i.e., collection date, locality, host plant, GPS coordinates, and collector name) of each sample are provided in Table S1. All collected specimens were deposited at the King Saud University Museum of Arthropods (Acarology section), Department of Plant Protection, College of Food and Agriculture Sciences, King Saud University (KSU), Riyadh, Saudi Arabia.  [37], collected from Egypt, Mexico, Pakistan and Saudi Arabia and seven host plants (Alnus jorulensis, Conocorpus erectus, Mangifera indica, Punica granatum, Rosa sp., Unknown and Vitis vinifera) in the present study (ARCGIS 10.5., esri.com computer software, Esri TM , Redlands, CA, USA).
Both adult females and males were mounted on glass slides in Hoyer's medium under a SZX10 stereomicroscope (Olympus, Tokyo, Japan). Samples were dried in an oven at 30-40 • C for five days. Male aedeagi from the punicae complex samples were imaged using an auto-montage software system (Syncroscopy, Cambridge, UK) attached to a DM2500 phase contrast microscope (Leica, Wetzlar, Germany) and then drawn using Adobe Illustrator software (Adobe Systems Inc., San Jose, CA, USA).

Morphological Study
Morphological investigations were performed using a BX51 fluorescence microscope (Olympus, Tokyo, Japan) in the Acarology Laboratory, KSU, for all 40 spider mite samples either collected from Egypt, Mexico, Pakistan, or SA (Table S1). In fact, the samples from Egypt (O. mangiferus), Mexico (O. punicae), and Pakistan (O. mangiferus) were initially identified/labelled and sent to us by local Acarologists [37,38] and our colleagues at our request.
Slides with mounted female and male (both dorsal and lateral position) specimens were comprehensively investigated. Diagnostic keys were used to identify the genus [39], subgenus, species group, and subgroup [1]. For the species-level identification, various regionally prepared diagnostic keys were consulted [3,4,[11][12][13]16,17,[40][41][42][43], as well as a diagnostic key that was prepared recently for the identification of world Oligonychus species [44]. However, due to observed inconsistency and ambiguity in aedeagal characters that were used in previously published keys/descriptions, differentiation among members of the punicae complex (O. punicae, O. mangiferus, and O. vitis) was found to be unreliable. Therefore, to investigate the actual taxonomic status of the punicae complex, a formula was devised by measuring the aedeagal parameters/characters, e.g., length of the shaft dorsal margin (L, measured from the base of the shaft dorsal margin to the level at which the shaft axis line crosses the shaft dorsal margin, the point at which the shaft dorsal margin starts bending downward), the width of the shaft (W, maximum width of the shaft measured near the base of the shaft dorsal margin), the height of the bent aedeagal part (H, measured from the level of the shaft ventral margin to the tip of the bent part), and angle (α) formed between the shaft axis and axis of the bent part ( Figure 3). Moreover, we comprehensively reviewed the taxonomic literature and compared all collected samples with the original descriptions of O. punicae [18], O. mangiferus [9] and O. vitis [19], and their subsequent re-descriptions that were made from different geographical localities [3][4][5]7,[10][11][12][13][14]17,25,41,45]. In addition, each sample was compared with the re-description (diagnosis) of O. punicae that was based on the type specimens (syntypes), which were observed by Dr. Jennifer J. Beard [46]. In addition, we tried to obtain and observe the type specimens of O. mangiferus and O. vitis with the help of acarologists in Pakistan and Egypt, respectively.
The types of O. mangiferus and O. vitis could not be observed in the present study, because either the information about the deposition of type specimens was missing in the original description of O. mangiferus [9], or types were lost as we were informed for O. vitis. However, eight samples (voucher no. 48, 71, 72, 73, 74, 75, 184, and 185; Table S1) from Pakistan including the specimens from the exact locality whence the original type of O. mangiferus was previously collected and described for the first time [9], and five samples from Egypt that were sent as O. mangiferus (voucher no. 42, 178, 180, 181, and 183; Table S1) were successfully analyzed. We also requested samples from Indian and Iranian acarologists, but did not receive the specimens of O. punicae from India (its type locality) and Iran.  (Table S1) of the punicae complex using a DNeasy mini kit (Qiagen, Hilden, Germany), following the manufacturer's guidelines. Six samples (voucher no. 75, 105, 178, 180, 181, and 183; Table S1) were not included in the molecular analysis due to the unavailability of sufficient specimens (Table S1). The concentration of the total genomic DNA solutions was assessed by a NanoDrop TM One spectrophotometer (Thermofisher Scientific, Waltham, MA, USA). DNA samples were stored at −20 • C after the sample was labelled with the appropriate field information.
The ITS2-rDNA region was amplified from the 34 mite samples using PCR-primers, ITS2-forward (5 -GTCACATCTGTCTGAGAGTTGAGA-3 ) and ITS2-reverse (5 -GTARCCTCACCTRMTCTGAGATC-3 ) [29]. In addition, COI-forward primer (5 -TGATTTTTTGGTCACCCAGAAG-3 ) and COI-reverse primer (5 -TACAGCTC CTATA-GATA AAAC-3 ) [32] were also used to amplify the COI-mtDNA fragments from 17 of the mite samples (Table S1). The PCR reaction was performed in a 30 µL reaction volume con-taining 15 µL 2× master mix (Molequle-On, Auckland, New Zealand), 0.4 µL of each 10 µM primer, 2 µL template DNA, and 12.2 µL Nuclease free water from Promega. The PCR conditions were as follows: a denaturation step at 94 • C for 5 min, followed by 35 cycles of a denaturation step at 94 • C for 60 s, an annealing step at 52 • C (for ITS2), and 53 • C (for COI) for 90 s and an extension step at 72 • C for 60 s, and a final extension step at 72 • C for 10 min. The PCR products were assessed in 1.2% agarose gel stained with acridine orange dye in 1× TAE buffer. Gels were observed and picturized using a gel documentation system (Uvitec, Cambridge, UK). For DNA sequencing, the PCR products were then purified using a Molequle-On PCR product purification kit (Molequle-On, Auckland, New Zealand).

DNA Sequencing and Analysis
The purified products of the ITS2 and COI regions were directly sequenced using the same primers at the Macrogen sequencing facility (Macrogen Inc., Seoul, Republic of Korea). A total of 34 ITS2 and 17 COI sequences were obtained from spider mite samples of the punicae complex, representing different locations across four countries (Table S1). The sequences were cleaned and edited using Bioedit software [47]. The cleaned sequences were searched against the NCBI GenBank database using BLASTn. Based on the BLASTn results, homologous and closely related ITS2/COI sequences were retrieved from GenBank and aligned with their counterpart sequences obtained during the present study, using the CLUSTALW multiple alignment tool in Bioedit.  (Table S1).

Phylogenetic and Genetic Distances Analysis
Phylogenetic analyses were conducted to assess the genetic variations within and among different samples of O. punicae, O. mangiferus, O. vitis (the punicae complex), and their closely related Oligonychus species using MEGA-X [48]. Phylogenetic trees were constructed using the neighbour-joining (NJ) and the maximum likelihood (ML) methods of the Tamura-Nei model [49]. The robustness of the tree branches was tested with 1000 replications in a bootstrap analysis [50]. The ambiguous positions in the nucleotides were removed for each sequence pair using the pairwise deletion method. Furthermore, the pairwise p-distances (intraspecific and interspecific genetic divergence) were also calculated using MEGA-X [47].

Morphological Analysis
The morphological investigations did not reveal any distinct differences in males and/or females from the 39 mite samples within the punicae complex (Tables S1 and S2), regardless of the collection localities across three countries (Egypt, Pakistan and SA) and six host plants (C. erectus, P. granatum, M. indica, Rosa sp., unknown host and V. vinifera). All aedeagal morphological parameters (e.g., L, H, W and α; Figure 3) were quite similar, regardless of whether they were collected from Egypt (Table S2; Figure 4D), Pakistan (Table S2; Figure 4E,F), or SA (Table S2; Figure 4A-C,G-H). Moreover, the male aedeagus of the specimens of O. mangiferus (voucher no. 71, Tables S1 and S2; Figure 4E) collected from the exact locality whence the original type was previously collected and described for the first time [9] almost resembled the original description/illustration of O. punicae [18], and did not show any differences. Exceptionally, only the male aedeagus of the claimed Mexican O. punicae (voucher no. 52, Tables S1 and S2; Figure 5A-D) was consistently different from the other 39 samples (Tables S1 and S2; Figure 4); e.g., the bent aedeagal part (H), was sub-equal, equal or longer than the length (L) of the shaft dorsal margin (vs. one-quarter to less than three-quarters the length of the shaft dorsal margin; Table S2; Figure 4), and the bent aedeagal part was approximately 1.8 to 2.5 times longer (H) than the shaft width (W) (vs. 0.7 to 1.6 times longer than the shaft width, in all other 39 samples; Table S2; Figure 4).

Molecular Analysis
The estimated pairwise nucleotide p-distances for the ITS2 sequences within the punicae complex ranged from 0.000 and 0.021 (0% to 2.1%) (Table S3). However, the nucleotide p-distances of the ITS2 between the punicae complex and their counterparts in the claimed Mexican O. punicae ranged between 0.172 and 0.188 (17.2% to 18.8%), indicating that the Mexican specimens are not closely related to the punicae complex (Table S3).
According to the ITS2-based NJ and ML phylogenetic trees (Figure 6), 37 sequences (samples) from the punicae complex from five countries (Egypt, India, Israel, Pakistan and SA) clustered together as a separate monophyletic clade with 87% and 94% bootstrap values, respectively ( Figure 6). The sequences of the punicae complex were further separated into four haplotypes, namely H1 (Egypt, Israel, Pakistan and SA), H2 (Israel), H3 (India), and H4 (India) (Figure 6; Table S3). Interestingly, the two sequences of claimed Mexican O. punicae were distantly located at the tree bases ( Figure 6).  Table S1), which needs to be re-identified. The genetic distances using the COI also showed low divergence (from 0.000 to 0.010; 0% to 1%) among members of the punicae complex from Egypt, India, Pakistan, and SA (Table S4). In addition, there was a low genetic divergence between O. vitis (Accession no. MW517748; India) and the other two members (O. punicae, O. mangiferus) within the punicae complex (Table S4). However, the nucleotide p-distances of the COI between the punicae complex and their counterparts in the claimed Mexican O. punicae ranged between 0.092 and 0.102 (9.2% to 10.2%), indicating that the Mexican specimens are not closely related to the punicae complex (Table S4).
In the present study, the COI-based molecular data revealed that O. vitis could be another junior synonym of O. punicae. The retrieved COI sequence of O. vitis (GenBank accession no.: MW517748, India) showed very low genetic divergence (0.3% to 0.6%) when compared with the other 19 COI sequences of O. punicae/O. mangiferus from Egypt, India, Pakistan, and SA. This limited variation falls within the range of intraspecific genetic divergence, as previously observed in various tetranychid species [28,30,33,35]. This sequence/sample of O. vitis was collected from the same location in India [55] as the type locality of O. punicae [18].
In the original description of O. vitis [19], the authors compared it with a re-description of O. mangiferus from African samples [51]. Although the male of O. vitis was described, the remarks were mostly based on the female holotype [19]. Additionally, it is questionable whether the original aedeagus illustration ( Figure 10A) of O. vitis was drawn in the proper lateral position [19], causing the diagram to show greater dissimilarity from the later redescriptions/illustrations ( Figure 10B-E), either reported from Africa [3,7] or India [13,17]. Meyer [3] differentiated O. vitis from O. mangiferus/O. punicae based on two aedeagal parameters, namely the angle (α) formed between the bent part and shaft, and the relative height of the bent part (H) to the length (L) of the shaft dorsal margin. Our observations showed that these two aedeagal parameters are variable within the punicae complex, and provide insufficient detail to separate O. mangiferus/O. punicae from O. vitis. This conclusion is supported by the various samples of the punicae complex ( Figure 4; Table S2) when compared with the previously described aedeagus for O. punicae ( Figure 10; Table S5a), O. mangiferus (Figure 9; Table S5b), and O. vitis ( Figure 10; Table S5c).
In addition, the molecular data of ITS2-rDNA and COI-mtDNA confirmed that the Mexican cryptic Oligonychus sp. does not belong to the punicae complex (Figures 6 and 7; Tables S3 and S4). The obtained interspecific range of genetic divergence of ITS2 and COI (17.2% to 18.8%, and 9.2% to 10.2%, respectively) that was detected between O. punicae/O. mangiferus/O. vitis (from Egypt, Israel, India, Pakistan, and SA) and the Mexican cryptic Oligonychus sp., is in accordance with separate species in previous molecular studies on tetranychid mites [28,29]. The ITS2-based interspecific nucleotide divergence ranged from 4.4% to 54.8% and has been observed previously in various tetranychid species including the three Oligonychus species of O. afrasiaticus, O. perseae, and O. mangiferus [29], whereas the COI-based interspecific sequence divergence among the 17 different Oligonychus species ranged from 7.3% to 18.3% [28]. In the present study, the interspecific nucleotide divergence ranged from 12.2% to 13.8% (ITS2 , Table S3) and 6.1% to 11.8% (COI , Table S4) between/among different closely related Oligonychus species, excluding the members of the punicae complex.
Additionally, based on our molecular analyses of DNA sequences of ITS2 and COI available in GenBank, O. punicae, O. coffeae, and O. ununguis sequences seem doubtful. For example, the COI sequence (Accession no.: KY474209; Figure 7) of the claimed O. punicae from the USA is highly divergent (8.3%, Table S4) from the claimed Mexican O. punicae (voucher no. 52, Table S1; Figure 7), which we propose as a cryptic Oligonychus sp. Similarly, the ununguis complex (ITS2, Table S3; Figure 6) and the coffeae complex (COI, Table S4; Figure 7) need further work to be resolved. The ITS2 sequence (Accession no.: HQ709242) of O. ununguis from China is highly divergent (11.5%; Table S3) from the ITS2 sequence (Accession no.: JF774179) of O. ununguis from Korea. In addition, the COI sequence (Accession no.: AB683671) of O. coffeae from Japan is highly divergent (8.9%; Table S4) from the COI sequence (Accession no.: KR870322) of O. coffeae from India.

Conclusions
In conclusion, the morphological and molecular data of the present study resolved the taxonomic problem of the punicae species complex by suggesting O. mangiferus and O. vitis as junior synonyms of O. punicae. Moreover, the obtained morphological and molecular data propose the claimed Mexican O. punicae [37] as a cryptic Oligonychus sp. that needs to be re-identified. The Mexican cryptic Oligonychus sp. should undergo a taxonomic revision as it is morphologically and genetically distinct from members of the punicae complex (O. punicae, O. mangiferus, and O. vitis). We also propose a revision of the Californian populations of the claimed O. punicae to confirm their taxonomic identity. Furthermore, we emphasize that the aedeagal parameters of the height of the bent part (H), the length of the shaft dorsal margin (L), and shaft width (W) are important aedeagal characteristics for delineating the closely related Oligonychus species. Finally, the present study highlighted the importance of the integrated taxonomic approaches in solving the problems related to species complexes. More studies should be conducted to resolve other species complexes of the genus Oligonychus, e.g., O. coffeae and O. ununguis.

Supplementary Materials:
The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/insects14010003/s1, Table S1: Geographical distribution, host plant, collection details, ITS2/COI fragment size and GenBank accession numbers of 40 spider mite samples of the punicae species complex collected from Egypt, Mexico, Pakistan & Saudi Arabia and analyzed morphologically/molecularly in the present study; Table S2: Relatively measured morphometric data obtained from different aedeagal parameters-viz. height of bent aedeagal part (H), length of shaft dorsal margin (L), shaft width (W), and the angle formed between shaft axis and axis of the bent part (α) of different spider mite samples of the punicae species complex collected from Egypt, Mexico, Pakistan and Saudi Arabia, in the present study; Table S3: Genetic divergence (pairwise p-distance) based on ITS2 sequences, either obtained in the present study or retrieved from GenBank, among various spider mite samples, representing different populations of four closely related Oligonychus species, reported from different countries; Table S4: Genetic divergence (pairwise p-distance) based on COI sequences, either obtained in the present study or retrieved from GenBank, among various spider mite samples, representing different populations of 18 closely related Oligonychus species, reported from different countries; Table S5

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.

Data Availability Statement:
All necessary data is provided in this paper.