Comparative Study of the Phenotypic Structure and Photoperiodic Responses of Female Asian Ladybirds Harmonia axyridis (Pallas) (Coleoptera, Coccinellidae) from Moscow, Belgorod, and Sochi

—Insect invasions are often accompanied by changes in the phenotypic structure of the populations and in the photoperiodic responses that regulate seasonal cycles. The Asian ladybird Harmonia axyridis has been recently discovered in Krasnodar Territory, from whence it is now gradually colonizing Central Russia. Our comparative study of individuals collected in 2020 in Sochi, Belgorod Province, and Moscow Province shows that interpopulation diﬀ erences in their phenotypic structure are nonsigniﬁ cant. The proportion of the light morph succinea is 77–83%; the proportion of the most common dark morph, spectabilis , is 13–20%. Laboratory experiments show that females from all the studied populations exhibit only a weak photoperiodic response. Comparative analysis reveals that females from the more northerly Moscow and Belgorod populations have a somewhat greater tendency to diapause and slightly delay reproductive maturation.

2020. Also, these populations originated from diff erent climates.
Our fi rst aim was to quantitatively assess the phenotypic structure of the invasive populations of the Asian ladybird, which is characterized by rich polymorphism in elytral color pattern. Dozens of color morphs of H. axyridis are known, underpinned by multiple alleles in a single autosomal locus (Dobzhansky, 1924;Tan, 1946). In the Far East, the light-colored morph succinea predominates, whereas in Siberia, the dark-colored axyridis is the most abundant one. The phenotypic structure of the native populations is remarkably persistent with the frequencies of the more common phenotypes remaining stable for decades (Dobzhansky, 1924;Vorontsov and Blekhman, 1986;Kholin, 1988;Belyakova, 2012;Andrianov et al., 2018). In view of this, it would be quite interesting to quantify the phenotypic structure of the invasive populations.
Another avenue of research addresses photoperiodic responses, which, through the induction of diapause, synchronize the life cycle with seasonal fl uctuations in the environment (Danilevskii, 1961;Tyshchenko, 1977;Zaslavskii, 1984;Tauber et al., 1986;Vinogradova, 1991;Denlinger, 2002;Saunders et al., 2002;Saulich and Volkovich, 2004;Danks, 2007;Tougeron, 2019). As a rule, the seasonal cycles of insects are adapted to the local climate, and so dispersion outside the native range is accompanied by the corresponding changes in the main parameters of the photoperiodic response (Saulich, 1999). However, exceptions to this rule also do exist. For example, the broadscale invasion of the Asian ladybird from East Asia to Southern, Western, and Central Europe was accompanied not by adjustment but by attenuation of the photoperiodic response so that the main role in the control of the seasonal cycle was transferred to diet-induced diapause (Reznik et al., 2015).
The objective of this work was to answer two questions: (1) does the phenotypic structure of the populations vary and (2) does the tendency for weaker photoperiodic sensitivity persist as H. axyridis is further expanding its distribution in the European part of Russia?

MATERIALS AND METHODS
The work was carried out with three laboratory colonies of H. axyridis that originated from individuals col-lected in the wild in the fall of 2020. The Sochi population was founded by 125 adults collected from overwintering aggregations in a house in Katkova Shchel Village (Lazarevsky District of Greater Sochi, 43.9°N, 39.4°E) in the last third of November 2020. The Belgorod population descended from 133 adults collected from overwintering aggregations in a house in the town of Shebekino (Belgorod Province, 50.4°N, 36.9°E) in the second third of October 2020. The Moscow population originated from 184 adults, pupae, and late-instar larvae collected from trees and shrubs during an outbreak in the town of Ramenskoe (55.6°N, 38.2°E) and the village of Bolshie Vyazemy (55.6°N, 37.0°E), both in Moscow Province, in the second third of October 2020.
The phenotypic structure was only studied in fi eld-collected individuals. The adults were sorted by the elytral pattern and the following morphs were identifi ed: succinea (hereinafter referred to as SUC), conspiqua (CON), and axyridis (AXY) (Dobzhansky, 1924). To compare the populations of diff erent geographic origin by their phenotypic structure, a metric of population similarity was calculated as r = √p 1 q 1 + √p 2 q 2 + … + √p n q n , where p 1 , p 2 ... p n are morph frequencies in the fi rst population (or sample), expressed as fractions, and q 1 , q 2 ... q n are frequencies of the respective morphs in the second population (or sample) (Zhivotovsky, 1991). These calculations were performed with Microsoft Excel 2010.
Photoperiodic responses. Prior to the experiments, ladybirds were reared on wheat aphids Schizaphis graminum Rond. (Hemiptera, Aphididae) at a temperature of 20-25°C and under a daylength of 18 h. The insects were thus maintained for 2 or 3 generations in the Laboratory of Biomethod of the All-Russia Institute of Plant Protection. The experiments were carried out in the Laboratory of Experimental Entomology of Zoological Institute (Russian Academy of Sciences). All the experimental larvae were kept at 25°C under a 14-h light and fed on peach aphids Myzus persicae (Sulz.), which in turn were reared on germinated beans of Vicia faba L. Ladybird pupae were kept under the same conditions. Newly emerged adults, no more than 24 h old, were transferred to plastic Petri dishes 60 mm in diameter and 15 mm tall and allocated among the following photoperiodic regimens (light:dark, h): 10L : 14D, 12L : 12D, 14L : 10D, and 16L : 8D. The temperature of 25°C was the same for all photoperiods.
During the experiments, adult beetles were fed on frozen eggs of grain moth Sitotroga cerealella (Oliv.) (Lepidoptera, Gelechiidae), which were glued onto a piece of cardboard by means of 30% sucrose solution. Also, in each Petri dish there was a plastic tube fi lled with water and stoppered with a cotton plug. Relative air humidity was the same in all the regimens (70%) and food was provided ad libitum. All the Petri dishes were examined daily and the date of the fi rst oviposition was recorded. At the end of the experiment (after 20 days) all the females that had not started egg-laying were dissected. This female age for dissection was chosen based on our previous studies where most nondiapausing females had started producing eggs by the 20th day under long-day conditions (Reznik et al., 2015). During dissection, the condition of the ovaries and that of the fat body were assigned a rank out of four as follows.
Stages of fat body development: 1-undeveloped (fat body transparent, almost indistinct; the viscera and the inner surface of abdominal tergites clearly visible); 2-weakly developed (fat body tissue represented by small globules that partially fi ll the body cavity); 3-mid-developed (fat body tissue sheath-like, consisting of globules that form numerous lobes); and 4fully developed (fat body tissue consisting of large globules with inclusions, occupying the whole abdominal volume, the viscera completely hidden in the folds of the fat body).
In total, 689 females were tested in the experiments (no less than 50 per population per photoperiod). Individuals that died during the experiments were discarded. Statistical processing of the results included correlation analysis, analysis of variance, probit analysis, and chisquared test. All computations were performed with SYSTAT 10.2 software.

RESULTS AND DISCUSSION
Phenotypic structure. There were no signifi cant differences in the phenotypic structure between the samples tested. Pairwise comparisons yielded similarity coeffi cients that ranged from 0.994 to 0.998. In all the populations (Sochi, Belgorod, and Moscow), the light-colored morph SUC predominated at frequencies of 77-83% (Fig. 1). Among the dark-colored morphs, SPC was the most abundant (13-20%). The frequency of CON never exceeded 2-4%. The morph AXY, which is rare in Europe, was represented by a single specimen from the Belgorod population.
These data suggest that, in terms of phenotypic composition, the Belgorod, Sochi, and Moscow populations are similar to invasive Western European ones as well as to native Far Eastern populations, the latter being the initial source of invaders (Lombaert et al., 2010). In Western Europe, the frequency of light-colored SUC individuals varies from 70 to 97%, irrespective of climate, season (sampling date) or biotope parameters Brown et al., 2008;Honěk et al., 2020). Similar frequencies of morphs are reported from remote regions with diff erent climate, e.g., Spain, Great Britain, and Italian Alps, where SUC constitutes 78-79% of the population (Honěk et al., 2020). In Belarus (Grodno, Brest Province, and Minsk), SUC strongly predominates at 96-97% (Kruglova et al., 2020), whereas, in adjacent Belgorod Province of Russia, according to our data, the frequency of this morph is lower at 83% (Fig. 1). In the southeast of the European distribution range of H. axyridis, the frequency of SUC is 70-80% in Kiev (Nekrasova and Titar, 2016), 80-89% in Crimea (Zakharov and Romanov, 2017), about 80% in the lowland of Krasnodar Territory (Orlova-Bienkowskaja, 2014), and, according to our results, 82% in Sochi (Fig. 1). Overall, the macrogeographic variability of the phenotypic composition of H. axyridis populations across its invasive European range is greater than in the native Far Eastern populations where SUC steadily predominates at 85-90% for decades (Kholin, 1990;Belaykova, 2011).
Photoperiodic responses. Preliminary analysis of the results of dissection showed that the degree of ovary development was negatively correlated with the degree of fat body development (Spearman's r = -0.898, N = 689, p < 0.001), as expected. For purposes of further analysis, all the females were divided into four groups (Table 1). Females with undeveloped or weakly developed ovaries and mid-or fully developed fat body were considered as being in diapause. All the females with fully developed ovaries (which were in the majority) were considered as being reproductively active. Females with mid-developed ovaries were regarded as being in the intermediate (transitory) condition. Finally, occasional individuals with both the ovaries and fat body undeveloped or weakly defi ned were deemed as being maldeveloped, starved or unhealthy and thus were excluded from analysis.
Probit analysis of the whole dataset (Table 2) showed that photoperiod signifi cantly infl uenced the diapausing-to-active female ratio. As the regression coeffi cients indicate, the fraction of diapausing females was lower and that of active females higher under long-day conditions as compared with short-day conditions. However, the photoperiodic responses that controlled reproductive  activity were relatively weak in females from all of the populations studied (Fig. 2). The fraction of reproductively active individuals signifi cantly increased under long-day conditions in Belgorod (p = 0.006) and Moscow females (p = 0.034) only, while in the Sochi population, the signifi cance of this photoperiodic eff ect was slightly below the threshold (p = 0.073).
The interpopulational diff erences were also generally weak and only observed for the fraction of diapausing females: individuals from the northern population entered diapause slightly more often (the diff erence was marginally signifi cant). Separate analyses showed that, within each photoperiod, interpopulational diff erences were only signifi cant in the short-day regimen (10L:14D): the fraction of diapausing females was higher and that of active females lower in the Belgorod and Moscow populations than in the Sochi population (Fig. 2).
The vast majority (85%) of reproductively active females commenced oviposition during the experiment. A two-way ANOVA of the entire dataset (n = 429) showed that maturation time (measured from adult eclosion to the fi rst egg laid) highly signifi cantly depended on the photoperiod (F = 7.2, df = 3, p < 0.001). The differences between the populations were much less significant (F = 3.6, df = 2, p = 0.029) and the interaction of these factors was nonsignifi cant (F = 1.5, df = 6, p = 0.173). As can be seen in Fig. 3, females from Sochi generally maturated somewhat faster than those from Belgorod and Moscow, but the signifi cance of this difference in each particular photoperiodic regimen was below the threshold. Maturation time was shorter under long-day conditions in females from all the populations but one-way ANOVA confi rmed the signifi cance of this eff ect only in the Belgorod population (F = 7.6, df = 3, N = 126, p < 0.001), but not in the Sochi (F = 0.7, df = 3, N = 156, p = 0.533) or Moscow (F = 1.7, df = 3, N = 147, p = 0.181) ones.
Geographic variation in the properties of photoperiodically induced diapause across diff erent populations has often been the focus of study in numerous insect species from various orders. In most cases, there exist substantial diff erences that are correlated, to a varying extent, with climatic variables. Often, there is a clinal pattern in the parameters of the photoperiodic response. Studies on widespread insect species even led to a generalization that, with a 5° change in latitude, the critical photoperiod for diapause induction changes by an average of 1.5 h (Danilevskii, 1961;Tyshchenko, 1977;Zaslavskii, 1984;Tauber et al., 1986;Vinogra-dova, 1991;Denlinger, 2002;Saunders et al., 2002;Saulich and Volkovich, 2004;Danks, 2007).
A number of studies on the intraspecifi c variation in the photothermic control of diapause were carried out with coccinellid species (Hodek, 2012). For example, there is a dramatic diff erence in the photoperiodic response of female Chilocorus bipustulatus L. from Leningrad Province and Middle Asia (Zaslavskii, 1970). Substantial diff erences were also discovered between populations of Coccinella septempunctata L. that are much closer to each other: beetles from Honshu Island, as a rule, undergo summer diapause, which is never the case in slightly more northerly beetles from Hokkaido (Ohashi et al., 2003). A comparative study of photoperiodic induction of winter diapause in populations of Hippodamia parenthesis (Say) between 40 and 44°N in the USA also revealed substantial intraspecifi c variation (Obrycki, 2020). By contrast, no signifi cant diff erences in the photoperiodic response were found between the populations of Propylea quatuordecimpunctata (L.) from Canada, Turkey, and southern France (Obrycki et al., 1993). The northernmost and southernmost of the studied populations of H. axyridis (Moscow and Sochi, respectively) are located 1300 km apart, which corresponds to a 12° diff erence in latitude and an 8°C diff erence in annual mean temperature, but the diff erence in the fraction of active and diapausing females between these populations is relatively small. On the whole, females from all the three populations studied exhibit a weak response to photoperiod: under any photoperiodic regimen tested, more than half of the individuals do maturate over the course of the experiment (Fig, 2), and their maturation time is only weakly related to daylength (Fig. 3), whereas the previously studied native females from Siberia and the Far East maturate signifi cantly faster under long-day conditions as compared with short-day conditions Vaghina, 2011, 2013;Reznik et al., 2015).
The northward expansion of invasive H. axyridis populations and their adaptation to colder climates seem to be accompanied by extremely hard selection: according to some data, winter mortality among adult beetles amounts to 99.8% . Still, this invasive species shows almost no diff erences in photoperiodic responses over a broad climatic gradient, probably because its seasonal cycle remains relatively independent of daylength even at higher latitudes (Reznik et al., 2015). This would explain why our study discovered in the northern populations of the Asian ladybird only a minor increase in the tendency to enter diapause and only a slight delay in maturation.

ACKNOWLEDGMENTS
The authors are deeply grateful to B. A. Borisov (All-Russia Research Institute of Phytopathology, Moscow) and L.N. Bugaeva (All-Russia Research Institute of Biological Plant Protection, Krasnodar) who helped with collection of material and to L.S. Ramenskaya and T.Ya. Umarova for their help with experimental work.

FUNDING
The work was supported by Russian Science Foundation (grant no. 20-66-47010).

COMPLIANCE WITH ETHICAL STANDARDS
The authors declare that they have no confl ict of interest. All the applicable international, national, and institutional guidelines for the care and use of animals were followed. All the procedures performed in studies involving animals were in accordance with the ethical standards of the institution or practice at which the studies were conducted.

OPEN ACCESS
This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.