Improving the Knowledge on Distribution, Food Preferences and DNA Barcoding of Natura 2000 Protected Species Paracossulus thrips (Lepidoptera, Cossidae) in Romania

Simple Summary NATURA 2000 species, Paracossulus thrips is one of the locally distributed, least studied moth species and is currently considered an endangered species in several European countries, even though the risk factors and its biology and ecology are not well known. In our study, conducted in Transylvania, Romania, we confirm the host plant as Phlomis tuberosa and describe the habitat type of P. thrips; we also present four new populations and the first DNA barcoding sequences for the species. Our study provides baseline knowledge on the biology and ecology of P.thrips, relevant for conservation, and for establishing management measures. Abstract Paracossulus thrips (Lepidoptera, Cossidae) is one of the locally distributed and endangered species. In Europe, it is also one of the few protected moth species, through Annexes II and IV of the Council Directive 92/43/EEC, Annex II of the Bern Convention. To date, little is known about the biology and ecology of this species. Our study was conducted in Transylvania, Romania. Romania hosts some of the strongest populations of the species in the European region. As part of the study, we conducted field observations, vegetation analyses, and genetic analyses. In our paper, we show the habitat types where we encounter P. thrips in Transylvania and confirm Phlomis tuberosa as a host plant. Furthermore, a piece of important information for habitat conservation is given. In this paper, we present the eggs and larvae of P. thrips, the first DNA barcoding sequences, and four new populations of P. thrips in Romania. Our study provides baseline knowledge about the biology and ecology of P. thrips, which is important for conservation and establishing management measures.


Introduction
Paracossulus thrips (Hübner, 1818) (Lepidoptera, Cossidae) is a locally distributed, poorly studied moth species. It is currently considered an endangered species in several European countries, although the risk factors are not well known. In Europe, it is also one of the few protected moth species, through Annexes II and IV of the Council Directive 92/43/EEC [1,2], Annex II of the Bern Convention and, not evaluated species in IUCN Red List. In the Romanian Red List of Lepidoptera, it is marked as vulnerable (VU) [3]. In the Bulgarian Red List it is critically endangered (CR) [2]. In the European part of its range, most populations are found in Romania, followed by the populations in Ukraine [4]. Although P. thrips is widespread, the known populations are relatively isolated from one another and no other populational data is known.
The species thrips was attributed to the genus Paracossulus by [43], considering its placement in the genus Catopta to be incorrect. The arguments leading to the placement of thrips in the genus Paracossulus were morphological aspects such as the rami of the antenna and the structural features of the thorax sclerites [43]. Later, important details were shown in the structure of the male genital apparatus [41].
Morphologically, P. thrips is closer to the genera Parahypopta and Cossulus than Catopta [43]. However, a comprehensive phylogenetic study of the Cossidae family is needed for a better understanding of the taxonomic placement of the P. thrips. The European populations of the P. thrips were attributed to ssp. polonica [62] which was later synonymized with the nominotypical form [31]. The species thrips was attributed to the genus Paracossulus by [43], considering its placement in the genus Catopta to be incorrect. The arguments leading to the placement of thrips in the genus Paracossulus were morphological aspects such as the rami of the antenna and the structural features of the thorax sclerites [43]. Later, important details were shown in the structure of the male genital apparatus [41].
Morphologically, P. thrips is closer to the genera Parahypopta and Cossulus than Catopta [43]. However, a comprehensive phylogenetic study of the Cossidae family is needed for a better understanding of the taxonomic placement of the P. thrips. The European populations of the P. thrips were attributed to ssp. polonica [62] which was later synonymized with the nominotypical form [31].
In this article, we attempt to fill some of these informational gaps by reporting four new populations in Romania, confirming the host plant for the larvae in the investigated areas, and performing DNA barcoding analyses to establish the taxonomic position of the species from a molecular perspective for the first time.

Field Survey
For observations on the biology and ecology of the species, field studies were conducted, in 2021, in Jucu de Sus, in Natura 2000 ROSCI0295 Dealurile Clujului Est (Cluj County), 26 km from Cluj Napoca, where P. thrips has a stable and vigorous population. To identify new populations, we conducted observations in habitats with steppe vegetation in Natura 2000 Protected areas: ROSCI0238 Suatu-Cojocna-Crairât; ROSCI0210 Râpa Lechint , a and ROSCI0272 Vulcanii Noroioi of Pâclele Mari and Pâclele Mici.

The Identification of the Larval Host Plant
To establish the host plant species with certainty, we used larvae obtained from the eggs of a female captured in Jucu de Sus. We used a sample of 300 eggs/larvae. After hatching, the larvae received fresh roots of Artemisa absinthium, A. maritima, A. vulgaris (Asteraceae) and tubers and root of Phlomis tuberosa, Nepeta nuda, and Salvia verticillata (Lamiaceae). Plants were harvested from steppe habitats where the presence of P. thrips is certain. The rhizomes, tubers and roots were cut into slices and placed in the container with the newly emerged larvae so that the larvae could have access to each of the plant species mentioned above.

The Study of Adult Moths
For studying the adult moths, the specimens were lured using light traps. The light source was a fluorescent Hg vapor lamp of 2 × 160 W, a UV blacklight lamp of 2 × 18 W and 3 Led UV light traps 30 LEDs, 5W. The lured specimens were photographed and released. All light traps worked between 20:30 and 02:00. For each studied location, one to five individuals were collected and stored in the Zoological Museum of the Babes-Bolyai University, Cluj Napoca (Tables 1 and 2). Freshly emerged specimens were collected from Jucu de Sus and kept alive to determine the lifespan and behavior of the adults and to obtain eggs.

Vegetation Analysis
The relevés were realized by using a stratified sampling procedure according to the Braun-Blanquet method. The abundance-dominance notes specific to the phytosociological relevé [63] were replaced by the percentage cover of species, estimated as the horizontal projection of each species on the soil surface. The sampling areas were 25 m 2 (5 × 5 m). To complete the habitat-level plant species inventory, the relevé method was combined with the linear transect method [64]. The habitats were established according to the literature [65]. The analysis of ecological categories of the investigated phytocoenoses was carried out to indirectly highlight the stationary conditions of each site. Thus, the percentage of species belonging to each ecological category was calculated at the site level. The identified phytotaxa were ecologically characterized according to the literature [66].
2.5. Molecular Analysis 2.5.1. Sampling and Collection of Data Four sequences of P. thrips were obtained for this study. The samples were collected from 2 localities (Table 2). Legs were stored in tubes with 96% ethanol. The moths are stored in the Zoological Museum of the Babes , -Bolyai University as vouchers. The sequences were obtained at the Biodiversity Institute of Ontario, Canada. DNA isolation, PCR amplification, and DNA sequencing followed standard protocols [67,68].

Sequence Analysis
For comparison with the cladogram of [43], we mined sequences of the Palearctic Cossidae from BOLD and NCBI. The full information regarding them is found in Appendix A, Table A1.
Sequence alignment and calculation of genetic distances were done in MEGA X software. Bootstrap analysis (1000 replicates) and the neighbor-joining tree of the COI sequences (the Kimura-two-parameter was used) were also built in MEGA X. We used as the out-group a sequence of Noctua fimbriata (Lepidoptera: Noctuidae).

Biology
During the field observations conducted in the Natura 2000 Protected area ROSCI0295 Dealurile Clujului Est (Jucu de Sus), and laboratory survey, we obtained information on adults, eggs, larval behavior, and food plant preferences which represent baseline data on the biology and ecology of the species.
The adult flight period is about four weeks, depending on weather conditions. The flight period may begin from mid-July to mid-late-August. In warm years, the first adults appear as early as late June. Adults are on the wing between 9:30 and 11:30 pm.
According to our observations, adults have low dispersal ability due to heavy flight and short activity/night time. The lifespan of adults held in captivity was 4-5 days; they do not feed (Figure 2A-C). The female lays eggs using the 1.5-2 cm long ovipositor. The eggs are oval, with a larger circumference in the basal third, and are covered with a sticky layer that makes them adhere very well to the substrate ( Figure 2D). The available female from Jucu de Sus laid 324 eggs; thereby we assume that a female lays more than 300 eggs. The larvae fed exclusively on the tubers of Phlomis tuberosa (Figures 3A-E and 4A-D), although fresh Artemisia, Nepeta nuda and Salvia verticillata roots were also available. On the field, we found that the newly hatched caterpillars entered the P. tuberosa stem in the crown area at the base of the leaves that forms the basal rosette of the plant ( Figure 3E). The larvae fed exclusively on the tubers of Phlomis tuberosa (Figures 3A-E and 4A-D), although fresh Artemisia, Nepeta nuda and Salvia verticillata roots were also available. On the field, we found that the newly hatched caterpillars entered the P. tuberosa stem in the crown area at the base of the leaves that forms the basal rosette of the plant ( Figure 3E).

Vegetation Description
Natura 2000 protected area ROSCI0295 Dealurile Clujului Est (Jucu de Sus) exhibits a slightly rugged relief, churned up by landslides. The dominant vegetation consists of meadow species interspersed with strips of Eleagnus angustifolia planted to stabilize the soil, alongside shrubs of Prunus spinosa, Crataegus monogyna and Rosa canina. P. tuberosa

Vegetation Description
Natura 2000 protected area ROSCI0295 Dealurile Clujului Est (Jucu de Sus) exhibits a slightly rugged relief, churned up by landslides. The dominant vegetation consists of meadow species interspersed with strips of Eleagnus angustifolia planted to stabilize the soil, alongside shrubs of Prunus spinosa, Crataegus monogyna and Rosa canina. P. tuberosa grows in the immediate vicinity of shrubs and has a high density, over 1000 stems. The meadow is used as sheep pasture all year round [69] (Figure 5A,B).
In the study area, a part of the meadow area is annually mown and the other part is set on fire. The vegetation structure includes steppe and forest-steppe species, with Stipa lessingiana, Stipa pulcherrima, Stipa tirsa and Stipa capillata dominating. The population of Phlomis tuberosa is located at the base of the slope in an area shaded by a Robinia pseudoacacia plantation and, the meadow is covered by ecotone species characteristic of the former xerothermic forests in the Transylvanian Basin. About 200 stems of P. tuberosa were observed in this area. Analyzing the vegetation, we assume that the surface occupied by P. tuberosa was much larger, but was restricted due to the invasion of Robinia pseudoacacia.
In the study area, a part of the meadow area is annually mown and the other part is set on fire. The vegetation structure includes steppe and forest-steppe species, with Stipa lessingiana, Stipa pulcherrima, Stipa tirsa and Stipa capillata dominating. The population of Phlomis tuberosa is located at the base of the slope in an area shaded by a Robinia pseudoacacia plantation and, the meadow is covered by ecotone species characteristic of the former xerothermic forests in the Transylvanian Basin. About 200 stems of P. tuberosa were observed in this area. Analyzing the vegetation, we assume that the surface occupied by P. tuberosa was much larger, but was restricted due to the invasion of Robinia pseudoacacia.
For the next two sites, we had the existing information in the standard site forms or management plans: The Natura 2000 protected site ROSCI0210 Râpa Lechința is located on the eastern bank of the river Mureș, near Lechința, and preserves a mosaic of steppe meadows formed on clay-rich and slightly saline soils. The site has an area of 283 hectares and is famous for some rare Lepidoptera species such as Cucullia mixta lorica, Hadula dianthi hungarica and Conisania poelli ostrogovichi. The habitat is 6240* Sub-pannonian steppe grasslands [71,72] ( Figure 5E,F).
The Natura 2000 protected area ROSCI0272 Vulcanii Noroioși of Pâclele Mari and Pâclele Mici is located in SE Romania, in the outer part of the Eastern Carpathians [73]. The nature reserve covers an area of 4 ha and is an area of geological and floristic interest, where several protected plant species can be found: Crambe tataria, Iris aphylla ssp. hungarica, Artemisia santonicum, Atriplex tatarica, Ephedra distachya, etc. The habitat is 1530* Pannonic salt-steppes and salt-marshes [74,75] (Figure 5G,H).

New Populations Identified in Romania
Following the field survey, we discovered four new populations of P. thrips in Romania:  Protected area ROSCI0238 Suatu-Cojocna-Crairât. Near Ploscoș (Valea Florilor) on Dealul Gorgan, four males of P. thrips were observed in 7 July 2015 by Sitar C. and Crișan A. The identified habitats in ROSCI0295 Dealurile Clujului Est (Jucu de Sus) are 6210 Semi-natural dry grasslands and 557 Scrubland facies on calcareous substrates (Festuco-Brometea). In ROSCI0238 Suatu-Cojocna-Crairat-Cojocna, the identified habitat is 6240 * Sub-558 pannonian steppe grasslands. The vegetation structure for both study areas is presented in Appendix A, Table A2.
For the next two sites, we had the existing information in the standard site forms or management plans: The Natura 2000 protected site ROSCI0210 Râpa Lechint , a is located on the eastern bank of the river Mures , , near Lechint , a, and preserves a mosaic of steppe meadows formed on clay-rich and slightly saline soils. The site has an area of 283 hectares and is famous for some rare Lepidoptera species such as Cucullia mixta lorica, Hadula dianthi hungarica and Conisania poelli ostrogovichi. The habitat is 6240* Sub-pannonian steppe grasslands [71,72] ( Figure 5E,F).
The Natura 2000 protected area ROSCI0272 Vulcanii Noroios , i of Pâclele Mari and Pâclele Mici is located in SE Romania, in the outer part of the Eastern Carpathians [73]. The nature reserve covers an area of 4 ha and is an area of geological and floristic interest, where several protected plant species can be found: Crambe tataria, Iris aphylla ssp. hungarica, Artemisia santonicum, Atriplex tatarica, Ephedra distachya, etc. The habitat is 1530* Pannonic salt-steppes and salt-marshes [74,75] (Figure 5G,H).

New Populations Identified in Romania
Following the field survey, we discovered four new populations of P. thrips in Romania:

Sequence Analysis
The four sequences of P. thrips were submitted to GenBank; they are the first published sequences of this species (accession numbers OK314991, OK314992, OK314993, OK314994). The sequences belong to individuals from two populations: three from Jucu de Sus (Cluj County) and one from Pădurea Babadag (Tulcea County) ( Table 2). The two populations are 450 km apart and are geographically isolated by a natural barrier (Carpathian mountains) ( Figure 1). The genetic distance between individuals is 0 and 0.002 (Table 3). The values of genetic distance between the sequences of P. thrips and the out-group are 0.6857 and 0.6846, respectively (Table 3). Full sequences of P. thrips are available in the Appendix A, Table A3.
In the molecular analysis, we examined 21 Palearctic species of Cossidae. Our dataset consisted of 191 sequences mined from BOLD and NCBI (see the electronic Supplementary Materials, Annex S1). The neighbor-joining (NJ) tree was made based on 113 of the sequences (Figure 8) for comparison with the cladogram of [43]. The lowest values of the recorded genetic distances are 0.0041 and 0.0166 between species Yakudza vicarius and Eogystia hippophaecolus, respectively, 0.0622 between Streltzoviella insularis and Kerzhnerocossus tannuolus (see the electronic Supplementary Materials, Annex S2). The nearest species to P. thrips is Eogystia sibirica (0.1036) (Figure 9). The matrix with the genetic distances between P. thrips and the species from the same clade is presented in Table 3. Catopta griseotincta is the most distant species from P. thrips (0.1862), after the out-group ( Figure 9). All genetic distances between specimens of P. trips and specimens of the genus Catopta are presented in (Table 4). The average genetic distance of the entire analyzed group is 0.1231 (without out-group) and 0.1328 (with out-group). Genetic distances for all sequences are available in the electronic Supplementary Material, Annex S2.
Our results show that DNA barcoding worked well in discriminating P. thrips sequences. Our results also show that DNA barcoding worked well in discriminating the 21 species considered in the analysis. Our results show that DNA barcoding worked well in discriminating P. thrips sequences. Our results also show that DNA barcoding worked well in discriminating the 21 species considered in the analysis.

Discussion
P. thrips is susceptible to extinction as a result of the geographical isolation of extant populations. There is also a lack of information on larval host plants, types of habitats, or population densities and dynamics. As such, this species is given protected status in Europe. In the first part of the study, we attempted to establish with certainty the larval host plant. To date, most authors consider Artemisia sp. to be the larval host plant [1,[5][6][7][8][9][10]. Ref. [4] states that the host plant is P. tuberosa, but there are no published studies to confirm this so far. Our studies confirm that P. tuberosa is the larval host plant for P. thrips in Romania The limited mobility and short flight time of adults also limit the colonization of new habitats. Thus, any pressure exerted on the remaining habitat by external factors will also affect the populations of P. thrips. For this reason, we consider the knowledge of the ecological preferences of P. tuberosa and population dynamics to be extremely important in order to better formulate any future management and protection implications.
Phlomis tuberosa is a continental Eurasian species. In Romania, it has been reported mainly in stands with woody vegetation, such as forests, bushes, vineyards, but also in hay meadows. It is often found in forest-steppe regions [76]. It occurs in stands with slight or no human impact, being considered oligohemerobic to mesohemerobic [77]. From an ecological point of view, the species is associated with xerothermic oak and deciduous forests, which are conditioned azonally by the orographic and edaphic particularities (Quercetalia pubescentis Br.-Bl 1931 em.), as well as with xeric grassland communities, established on cleared slopes in the oak area [66].
Thorough knowledge of habitat preferences can provide the data needed to identify potential habitats, including the identification of any new populations of P. thrips. Although most of the moth populations in the European part of the range are on the territory of Romania, the distribution of the species in this country is not yet fully characterized. For example, the presence of the nearby Hungarian populations from Hortobágy and Gyula [4] possibly indicates the existence of a population of P. thrips in the western part of Romania.
Although the light traps worked for almost 6 h, between 20:30 and 02:00, all individuals were observed between 09:30 and 10:30. The short flight interval, the isolated populations, and the direct observations made on the field indicate a low dispersal capacity of the individuals. Another limiting factor that contributes to the reduced dispersion is the short life span of adults. This aspect is also characteristic of other species of Cossidae [78] due to the atrophied buccal apparatus of adults that do not feed.
The second part of our study attempted to confirm the current taxonomic placement of P. thrips. Until recently, P. thrips was classified as Catopta [62], but in 1990 the species was included by [43] in the distinct genus Paracosulus. Ref. [43] presents a cladogram (based on the external morphology of the adults) that places the species P. thrips closer to the genera Dyspessacossus, Parahypopta, and Dyspessa. Furthermore, the genus Catopta is assigned to the subfamily Catoptinae [79] that was described based on the apomorphic features of the genitals of both sexes [79].
This framing is also supported by our DNA barcoding analysis. The sequences of P. thrips are grouped in the same cluster and form a clade with the species Eogystia sibirica, Parahypopta caestrum, Mormogystia proleuca, and Holcocerus gloriosus. All of them are grouped into a larger clade with the genus Dyspessa represented by three species: D. ulula, D. Psychidion, and D. salicicola. Thus our preliminary genetic data confirms the morphological studies of [43] and they argue the use of the genus Paracossulus instead of the genus Catopta. This is supported both by the position of the two genera in the phylogenetic tree and by the genetic distance between Paracossulus and Catopta which is on average 0.1832 (Table 4). A comparison of P. thrips with species of the genera Dyspesacossus and Cossulus was not possible due to the lack of available sequences.
Despite the geographical isolation, there are no significant differences between the sequences of P. thrips from the specimens from Jucu de Sus and the specimen from Babadag Forest (Figure 1) (Table 3). This lack of genetic diversity between the various isolated populations requires further studies with different methods. P. thrips and its habitat are directly threatened by overgrazing (especially by sheep and/or goats) at most or all sites, by mowing of meadows or by other land mismanagement practices which lead to rapid habitat degradation. As such, conservation measures for P. thrips are acutely needed to prevent these threats and management practices must be substantiated based on available data. There was a fine mosaic of over-exploited to abandoned places. All meadows as habitat were grazed and mown for hay and part of the steppe-like meadows were commonly burned. Habitat changes similar to those happening in the rest of Europe are present here too. Namely, meadows are changing to more homogenous ones, without the fine mosaic of micro-habitats. It seems that for a species such as P. thrips traditional burning (in spring, with cold nights) is not problematic.