Ligustrum lucidum W. T. Aiton (broad-leaf privet) demonstrates climatic niche shifts during global-scale invasion

Biological invasions are a major threat to global biodiversity. Ligustrum lucidum, native to temperate Asia, is one of the most invasive plant species in the world. Climate is an important ecological factor influencing species distribution. Therefore, we investigated the climatic niche of L. lucidum in various regions of the world to determine whether it uses different climatic conditions in its invasive ranges than in its native range. The geographical coordinates of its occurrence were extracted from the Global Biodiversity Information Facility and Southern African Plant Invaders Atlas databases. Climatic variables and altitude data were obtained from WorldClim. We evaluated niche overlap and performed niche similarity tests, and estimated niche shift parameters. L. lucidum occurs mostly in warm temperate climates. Niche overlap between native and invaded areas was low. Niche similarity tests indicated that the species could expand its occurrence into regions with climates similar to and different from that of its native range. We concluded that L. lucidum uses different realized climatic niches in its invasive ranges than in its native range. Warmer and wetter climatic conditions may not necessarily constrain this species from establishing populations outside of its native range.


Results
L. lucidum is present on all continents except Antarctica, occurring mostly in warm temperate climates (Köppen-Geiger climate type C) (Fig. 1). Even within the same macroclimate, there was low climatic overlap between the invaded regions of the plant and its native range. The D values varied from 0.00 (Southeastern South America) to 0.12 (Oceania) ( Table 1). Except for South Africa and the North American West Coast, the low D values reflected high proportions of expansion (E > 70%) and unfilled (U > 90%) climatic conditions. In South Africa and the West Coast of North America, the expansion values (E %) were relatively low (<35%). Therefore, the climates of these regions were highly similar to those of the native area.
According to the niche similarity test, however, only the North American West Coast has climatic conditions similar to those in the native range in China (Sim.test: p < 0.05). Therefore, the expanded climatic conditions (E %) in this region correspond to the same conditions observed in the native range. However, the climatic conditions of the other invaded areas were dissimilar to those observed in the native range (Sim.test: p > 0.05). The climates of the expansion areas (E %) differed significantly from those of the native range. In these cases, L. lucidum shifted its climatic niche during the invasion process.
For regions in which L. lucidum shifted its climatic niche ( Fig. 2a-g), the first two axes of the multivariate ordinations explained a large proportion of the total inertia (>60%). In general, L. lucidum expanded its occurrence to warmer and wetter areas (Oceania, Europe, and Southeastern South America) and areas with comparatively lower seasonality of precipitation (North American East Coast) and temperature (Central America and Mexico) than that of the native area.
On the North American West Coast, L. lucidum occupied the same climatic niche as its native range (Fig. 2e). In this area, expansion occurred under similar climatic conditions. The first two axes of the multivariate ordinations explained a large proportion of the total inertia (66.6%; Axis 1 = 45.2%, and Axis 2 = 21.4%). Figure 2e  35 and Ligustrum lucidum geographic distribution focusing on the area of its native occurrence in China and other invaded areas for which the climate niche dynamics were evaluated. This figure was generated using R 3.5.1, with code adapted from http://www. rforscience.com/portfolio/koppen-geiger/ 52 . www.nature.com/scientificreports www.nature.com/scientificreports/ shows that the climatic conditions of L. lucidum on the North American West Coast were mainly analogous (blue color) to those in the native distribution area.

Discussion
L. lucidum tends to be found within the same type of macroclimate, namely warm temperate 35 . Nevertheless, we showed that it can invade regions with climatic conditions both similar to and different from those found within its native range. Considering that our analysis was conducted using field data that included biotic interactions, this fact suggests that the native range of L. lucidum only represents a portion of the fundamental niche of the species. In this sense, our study supports the findings of other investigators who demonstrated the potential of species to use different realized climatic niches during invasion 16,36,37 .
A complex interaction among ecological-evolutionary forces such as biological associations, propagule pressure, and adaptive evolution processes 15,[38][39][40][41] can explain why an invasive species may not retain the realized niche from its native range. When introduced into a new community, an invasive species may be favored by the lack of negative biological interactions present in its native range and by the potential presence of new mutualists [12][13][14][15] . In addition, the repeated introduction of an invasive species may increase propagule pressure 39 and foster evolutionary processes that lead to invaders with more adaptive potential 40 . Moreover, an invasive species may be preadapted to conditions that are present in the invaded range but are no longer readily available in the native one 4 .
Based on our results and those previously reported, we inferred that the realized niche shifts and worldwide distribution of L. lucidum are the result of a biological invasion "perfect storm" created by intentional human distribution, positive biotic interactions, and strong abiotic plasticity. This species has been widely planted as an ornamental tree by humans [23][24][25][26][28][29][30][31] . In certain municipalities of southern Brazil, L. lucidum is the most abundantly planted urban tree 42 . Moreover, the fruit of L. lucidum may be avidly consumed by local birds 43,44 , which disperse tree propagules in natural areas where they grow rapidly under various ecological conditions 24,34,45 . It is, therefore, evident that L. lucidum is potentially a serious invasive species.
Overall, our results showed low climatic niche overlap (D) and expansion into areas possessing different climatic conditions (E %). L. lucidum expanded its presence into more humid and warmer areas with less temperature and precipitation seasonality. Therefore, the species may be tending towards the invasion of wet tropical areas. Tropical ecosystems constitute a vital part of global biodiversity 46 and the invasion of plant species such as L. lucidum could place even more pressure on these ecosystems 47 . Examples of these ecosystems can be found in regions in the Brazilian Atlantic Forests, Mesoamerican Forests, East Australian Forests, Madagascar, and the islands of the Indian Ocean. These observations are alarming when one considers biological invasion together with ongoing climate change. An invasive species may be able to thrive under many new environmental conditions in the future 48 .
In conclusion, on a global scale, L. lucidum has invaded and occupied regions with climates similar to and distinct from that of its native habitat. In Oceania, Europe, Southeastern South America, Central America and Mexico, the North American East Coast, and South Africa, L. lucidum thrives under various climatic conditions. In these areas, invasion occurred primarily along temperature and precipitation gradients, towards wetter and warmer areas with lower seasonality in temperature and rainfall. This indicates that the native range of L. lucidum only represents a portion of its fundamental niche. Considering the invasive potential of this species, policies are required to limit or restrict the use, sale, and transport of L. lucidum outside its native range. In this way, its establishment can be prevented in areas where it has not yet been introduced.  www.nature.com/scientificreports www.nature.com/scientificreports/  www.nature.com/scientificreports www.nature.com/scientificreports/ Considering the native and invaded range climates, principal component analyses (PCAs) were performed pairwise for each study region to investigate variations in realized climatic niche 53,54 . In this way, a two-dimensional climatic species distribution space was created. Ordinations were determined using kernel density estimation for species occurrence 21 . Niche overlap was determined according to Schoener's D index, which ranges from 0 (no overlap) to 1 (complete overlap).
Niche similarity tests were performed 54 (α = 0.05) by comparing observed and randomly simulated (n = 100) D values. For this test, an observed D value greater than the randomly simulated D values was considered a null hypothesis, meaning that the invaded niche resembled the native niche; therefore, there was no climate niche shift. We also determined expansion into climatic conditions that are distinct from those in the native range (E) and the unfilled climatic conditions in the native area of the species that are absent in the invaded area (U).

Data Availability
The data used in this research are available upon request by contacting the corresponding author.