New insights into virus yellows distribution in Europe and effects of beet yellows virus, beet mild yellowing virus, and beet chlorosis virus on sugar beet yield following field inoculation

Beet yellows virus (BYV), beet mild yellowing virus (BMYV), beet chlorosis virus (BChV), and beet mosaic virus (BtMV) cause virus yellows (VY) disease in sugar beet. The main virus vector is the aphid Myzus persicae . Due to efficient vector control by neonicotinoid seed treatment over the last decades, there is no current knowledge regarding virus species distribution. Therefore, Europe-wide virus monitoring was carried out from 2017 to 2019, where neonicotinoids were banned in 2019. The monitoring showed that closterovirus BYV is currently widely spread in northern Europe. The poleroviruses BMYV and BChV were most frequently detected in the northern and western regions. The potyvirus BtMV was only sporadically detected. To study virus infestation and influence on yield, viruses were transmitted to sugar beet plants using viruliferous M . persicae in quadruplicate field plots with 10% inoculation density simulating natural infection. A plant-to-plant virus spread was observed within 4 weeks. A nearly complete infection of all plants was observed in all treatments at harvest. In accordance with these findings, a significant yield reduction was caused by BMYV and BChV (−23% and −24%) and only a moderate reduction in yield was observed for BYV (−10%). This study showed that inoculation at low densities mimics natural infection, and quick spreading induced representative yield effects. Within the background of a post-neonicotinoid era, this provides the basis to screen sugar beet genotypes for the selection of virus tolerance/resistance and to test the effec-tiveness of insecticides for the control of M . persicae with a manageable workload. following purification using the NucleoSpin PCR Clean-up kit (Macherey-Nagel) following the manufacturer's instructions. The species assignment of the obtained sequences was carried out based on comparison to GenBank reference sequences for BMYV, BChV, and BWYV.

Vereinigung Zucker/ Verein der Zuckerindustrie, 2020). With the intensification of sugar beet cultivation, some pests and diseases became a major threat for sugar production, restricting the agronomic yield potential of the crop (Bennett et al., 1956). One of the most economically important viral diseases is virus yellows (VY), caused by a complex of different aphid-transmissible virus species. In Europe, beet yellows virus (BYV), beet mild yellowing virus (BMYV), and beet chlorosis virus (BChV) are most abundant, while beet mosaic virus (BtMV) is only rarely observed. The spread of beet western yellows virus (BWYV) is so far restricted to the USA and Asia and not relevant for the European sugar beet cultivation (Stevens et al., 2005;Xiang et al., 2008). Myzus persicae, the green peach aphid, is the main vector for all aphid-transmitted virus species in sugar beet fields (Limburg et al., 1997;Schliephake et al., 2000;Kozlowska-Makulska et al., 2009).

Beet yellows virus belongs to the genus Closterovirus in the family
Closteroviridae. Infections with BYV lead to yellowish discoloration of the older leaves and subsequently reddish necrosis may occur (de Koeijer and van der Werf, 1999). BYV can be transmitted by more than 20 different aphid species in a semipersistent mode of transmission. In addition to M. persicae, Aphis fabae can contribute to virus transmission (Limburg et al., 1997). BYV virions move via the phloem, but can also colonize mesophyll and epidermal cells and have been detected in plasmodesmata that connect the different cell types of the phloem (Dolja, 2003;Dolja & Koonin, 2013).

Beet mild yellowing virus, Beet western yellows virus, and Beet chlorosis virus belong to the genus Polerovirus in the family Luteoviridae. In
Beta species, the viruses induce yellow to orange leaf discoloration, which may cause premature foliage death (Lewellen et al., 1999).

Beet mosaic virus belongs to the genus Potyvirus in the family
Potyviridae. Symptoms of BtMV infection initially appear as yellowish speckles before the typical mosaic-like structures appear. In addition, the leaves are often malformed (Dunning & Byford, 1982). BtMV is transmitted via a nonpersistent mechanism (Gallet et al., 2018). The main vectors for the transmission of BtMV in the field are M. persicae and A. fabae (Dusi & Peters, 1999). However, the virus can also be transmitted by many other species such as Myzus ascalonicus (Semal, 1956), M. euphorbiae, Acyrthosiphon pisum, Metopolophium dirhodum, and Rhopalosiphum padi (Dusi & Peters, 1999).
Studies to investigate the occurrence and spread of VY species are rare. The most prominent study was published in 2005 and was based on a sampling of around 260 sugar beet leaves with symptoms in 10 countries on three continents (Stevens et al., 2005). BMYV was most frequently found in northern and western regions of Europe, BChV was also found in southern areas of Europe and Chile, and BYV was predominantly detected in southern Europe, Turkey, the USA, and Chile. Field studies to investigate the influence of infection with VY species on sugar beet yield have so far been carried out exclusively with an inoculation density of 100%, meaning that each plant was individually inoculated with at least 10 wingless M. persicae individuals (Smith & Hallsworth, 1990;Stevens et al., 2004). Infection with BYV led to yield losses of up to 47% (Smith & Hallsworth, 1990).
Infection with BMYV reduced yield up to 29% (Smith & Hallsworth, 1990) and was shown to result in an 18%-27% reduction in sugar yield. More varying losses in sugar yield of 8%-24% were observed for BChV (Stevens et al., 2004). The field studies showed that BMYV is more damaging to root and sugar yields when inoculated early in the growing season (May and June) compared to BChV, but when plants become infected later (July), BChV has a greater impact on yield (Stevens et al., 2004). BYV infection later than July has only minor effects on sugar beet growth (Smith & Hallsworth, 1990).
Yield loss induced by BtMV infection was not observed to be higher than 10% (Dunning & Byford, 1982). However, the studies conducted so far do not resemble natural infestation in the field. The disease starts from individually infected plants in the field, leading to so-called infection patches, and can spread to the entire sugar beet field if supportive conditions are present in the further course of the vegetation period.
As the yellowing viruses were satisfactorily controlled by combating the vector with insecticides from the neonicotinoid class since the early 1990s, VY disease lost its economic importance and working groups across Europe gave up their interest in monitoring the occurrence and distribution of the virus species involved. However, pesticides are coming into focus, concerning costs, safety, environmental impact, and the development of resistance in the target organisms (Luterbacher et al., 2004). Also of particular importance is that seed pelleting with neonicotinoids has been banned in sugar beet cultivation since 2019 in most of the sugar beet-producing countries in Europe. As a result, virus-carrying aphids can colonize the crop earlier in the vegetation period leading to high potential yield losses. Because M. persicae has already developed resistance to the remaining classes of insecticidal active substances, resistance breeding will offer the only practical solution for disease control in the future (Luterbacher et al., 2004).
It is not known how the occurrence of the yellowing viruses has changed after almost 30 years of effective vector control.
Hence, the aim of this study was to gain an up-to-date overview of occurrence and distribution of the VY species, concentrating on a number of European countries through countrywide virus monitoring. In addition, by applying an inoculation method with 10% inoculation density, the natural infection course of the disease in a field was simulated in order to determine the yield effects under viral infection. Effective genetic resistance to VY species is currently not available in sugar beet, reflecting the success of former pesticide usage for control and plant breeder's priorities, for example, higher yield and quality (Luterbacher et al., 2004).
The field inoculation experiments carried out in this study provide the basis for establishing screening tests for sugar beet genotypes in order to identify and select resistance/tolerance and to ensure yield security even under aphid infestation by growing resistant/ tolerant varieties.  Amplified products were sequenced directly following purification using the NucleoSpin PCR Clean-up kit (Macherey-Nagel) following the manufacturer's instructions. The species assignment of the obtained sequences was carried out based on comparison to GenBank reference sequences for BMYV, BChV, and BWYV.

| Determining impact of virus infection on sugar beet yield parameters under field conditions
In 2019, the test sites were located in Sieboldshausen and Wollbrechtshausen near Goettingen, Germany. Seeds of the susceptible cultivar Vasco (Sesvanderhave) were sown in 3-row field plots of 8 m length with 100 plants each in four replicates, by the end of March. The replicates for the two treatments, inoculated and non-inoculated, were in the same block, but were separated from each other by non-inoculated border rows. As inoculation variants, BMYV, BChV, and BYV were chosen. To investigate possible synergistic effects of yellowing virus species in plants, a coinoculation of BChV and BYV was also included in the experiment.

| Yield and quality analysis
Sugar beets from inoculated and non-inoculated plots were harvested by a sugar beet harvester by the end of October. Beets were washed and freed from leaf residues. Yield was determined by weighing the total plot for each repetition. Beet brei was prepared and analysed according to routine methods of the sugar industry

| Statistical analysis
For the field trial with virus inoculation, a completely randomized block design was not practicable because the potential risk for crosscontamination of the different virus species was too high. In order to avoid an unintentional mixing of the virus species, experiments were carried out at two different locations near Goettingen with a randomized design without blocks for each virus species. Because the effects of the location are difficult to estimate and yield data for the genotype used were missing, statistical differences between the different virus species in terms of yield reduction could not be calculated. The statistical analysis was limited to two samples, which were tested by a t test to determine the significant differences between virus-inoculated and corresponding non-inoculated plots in a fourfold repetition. Significant differences were determined using SigmaPlot and indicated by *p ≤ .05, **p ≤ .01, and ***p ≤ .001.

| Europe-wide monitoring to detect VY species in sugar beet
All species of the VY complex could be detected. BYV or the poleroviruses were predominantly found in all years, whereas BtMV was detected relatively rarely. The virus species were identified in 8 out of 10 European countries, with no virus detection at all in Hungary and Sweden; however, the results for Hungary were based on low overall sample numbers. Because the number of samples per country was not consistent and varied from one year to the next, the results are expressed as percentage of samples that tested positive (Table 1).
In 2017, a total of 3,091 samples were collected. In 683 samples, yellowing viruses could be detected, which represents a percentage of 27.9%. Within the infected samples, 18.9% tested positive for BYV, and 6.8% and 2.2% for the poleroviruses and BtMV, respectively. The country with the highest percentage of BYV-infected samples was Spain (63.3%), followed by the UK (56.8%) and France (30.9%). Lower percentages were found in Germany (16.6%), the Netherlands (7.3%), and Denmark (3.2%). In Belgium, Italy, Hungary, and Sweden, BYV was not detected at all. Countries with the highest percentage of polerovirus-infected samples were France and the UK (25.6% and 13.5%, respectively), followed by Germany (4.6%) and Italy (3.3%). In the remaining countries there was no evidence for polerovirus occurrence. BtMV was detected in the UK (24.3%), France (5.5%), Germany (1.3%), and the Netherlands (0.8%).
In 2019, in which the use of neonicotinoid seed treatment was prohibited in most of the European monitoring countries (except Belgium and Hungary), a total of 1,334 samples were collected. Yellowing virus species were detected in 480 samples, which represents a percentage of 28%. Among the infected samples, 9.7% were positive for BYV, 25.7% for the poleroviruses, and 0.5% for BtMV. The country with the highest percentage of BYV infection was Spain (91.2%), followed by the UK (35.5%). Lower percentages were observed for Italy, Germany, Belgium, France, and the Netherlands, ranging from 1% to 7.4%. The countries with highest percentages of polerovirus infections were France (67.3%), the Netherlands (40.8%), Belgium (26.7%), and the UK (25.8%), followed by Germany (18.9%), Spain (5.9%), and Italy (3.3%). No poleroviruses were detected in the remaining countries. BtMV was detected only in Spain (2.9%) and Germany (0.8%).

| Polerovirus discrimination and verification of virus infection
As the serological method applied did not distinguish between the different sugar beet-infecting polerovirus species, the assignment to  the species was carried out by sequencing (Table 2). In contrast, the combination of BYV and BtMV, and even triple infections with poleroviruses, were more common (data not shown).

| Symptom development after M. persicae inoculation and virus transmission in the field
First symptoms for BMYV, BChV, BYV, and the coinoculation of

| Effect of virus infection on root yield and white sugar yield
The root yield data (t/ha) are displayed in Table 3. The mean root yield in non-inoculated plots was 105 t/ha. This value is slightly above the average yield achieved in the same year on nearby sites, for   (Figure 3).

Non-inoculated
The mean WSY (t/ha) of inoculated and non-inoculated plots are summarized in Table 3. A reduced root weight after infection with the different VY species resulted in a corresponding reduction in sugar content, and accordingly WSY. Our data show that after virus infection, potassium was significantly reduced, while sodium as well as amino-N contents were significantly increased, leading to a deterioration of beet processability and hence reduction of the WSY (data not shown).
Infection with BMYV and BChV led to equal values in WSY (12 t/ ha). This was a reduction of 29% (p ≤ .0001) compared to control plots. Sugar beets inoculated with BYV showed a WSY of 16.5 t/ ha, which was a reduction of 11% (p ≤ .0023). The reduction in WSY was highest in the coinoculation treatment with BChV/BYV (9.7 t/ ha) compared to the average of non-inoculated controls (17.5 t/ha), which represented a reduction of 43% (p ≤ .0001; Figure 3).

| D ISCUSS I ON
The inexperienced. Thus, yellowing of leaves can be caused not only by viral infections, but also a variety of other biotic or abiotic factors, for example, soil compaction and low microbial activity in the soil can cause nutrient (magnesium or boron deficiency) and water stress. Insects like the capsid bugs from the Miridae family can also cause yellowing due to their sucking activity on the leaves (Draycott, 2008). Some fungal diseases caused by Fusarium or Verticillium species can also cause chlorotic tissue (Hanson & Jacobsen, 2009;Strausbaugh et al., 2016). Finally, herbicides can have toxic properties that can lead to leaf yellowing (Draycott, 2008). The occurrence of BYV did not follow a clear trend, but reflects the natural fluctuations that were reported from previous studies (Stevens et al., 2005). According to this virus monitoring, BYV seems most widely distributed in Spain. However, contrary to previous ob- In our study we could show that in some countries mixed infec- The field trial was designed to provide up-to-date information on the symptom development and potential yield reduction as well as quality effects under nearly natural conditions. In order to ensure this, a method with 10% inoculation density was established.
Considering that in nature only about 0%-8% of the M. persicae adults are actually viruliferous (Stevens et al., 1995), the application method of 10% inoculation density simulates a strong natural infection pressure, in contrast to the formerly used 100% inoculation density; however, it is still closer to the natural conditions in a field.
The study aimed to generate measurable yield effects even at lower inoculation densities than previously applied, primarily to reduce the amount of work required. In the case of transmission of M. persicae, one must also expect that some insects may be decimated by injuries, antagonists, or even wind and heavy rain. Each virus species in the VY complex differs in its effect on yield (Stevens et al., 2004;Wintermantel, 2005 Surprisingly, the yield losses in BYV-infected plots were comparatively low, which differs from previous studies. BYV was always described as the most economically relevant virus, which generates yield losses of up to 50% (Smith & Hallsworth, 1990). Clover et al. (1999) described the infection with BYV as markedly decreasing sugar yield, which is related to the reduction of storage root growth. Another study on leaf growth and leaf area index (LAI) showed that BYV caused a higher reduction in leaf light interception than BMYV, resulting from a lower LAI and a higher proportion of yellowed leaf area after BYV infection (de Koeijer & van der Werf, 1999). The mechanism by which BYV reduces sugar beet growth and root yield is unknown (Clover et al., 1999).
Nevertheless, these data are to be considered as preliminary results, as only one sugar beet variety, one location, and one year has been investigated so far. If our findings are confirmed, the relevance and order of importance of the different virus species may need to be reconsidered.
As already expected following the observations of strong symptoms and the early achievement of infection rates of 100%, the highest yield loss of about 40% occurred in the plots with the coinoculation of BChV and BYV. Due to the fact that in a single infection with BYV the yield decreases only moderately, we assume that poleroviruses might have greater relevance for reduction of yield and quality parameters. Furthermore, considering the preliminary data on coinoculation of BYV with BChV, showing that BYV causes serious yield losses and quality reduction, this may be due to synergism with other yellowing viruses. Therefore, it will be important to repeat such coinoculation experiments with different virus combinations to study the synergism effects in more detail.
With the ban on neonicotinoid seed treatment since 2019, sugar beet cultivation will face new challenges. Sugar beet seedlings are no longer protected, especially in the critical emergence phase. Aphids can colonize the plants earlier, partly because the changed climatic conditions mean that more adults, possibly viruliferous individuals, can survive the winter. This not only affects the time of first colonization, but consequently also the yield that can be achieved from infected fields at the end of the vegetation period. The effect of the remaining classes of insecticidal active substances may be severely limited in regions where M. persicae populations with resistance properties occur, so that control via crop protection must be assessed as very critical. In order to keep yields stable even after M. persicae infestation, it is essential to breed virus-tolerant/resistant varieties. The challenge of breeding will be that VY is not just one pest, but a number of viruses from different virus families are involved, and therefore solutions for overall control must be found via conserved resistance mechanisms. For this purpose, it is important to continue virus monitoring, as it is not yet possible to estimate how the occurrence of the virus species will change in the growing regions without the use of neonicotinoids. The breeding process should also be supported by the determination of aphid flight times and by the identification of viruliferous M. persicae, and also resistance properties against insecticides for spray application, in order to be able to achieve the goal of bringing tolerant/resistant varieties onto the market.
Artificial field inoculation on which this work is based on provides the prerequisite for supporting breeders in the selection of suitable breeding material. Field inoculation with an inoculation density of 10% is a good option in terms of time and work intensity for screening sugar beet genotypes under natural growing and inoculation conditions. This will enable farmers to ensure a secure harvest in the near future by growing virus-tolerant or virus-resistant varieties.

ACK N OWLED G EM ENTS
We would like to thank all cooperating European sugar beet breeders of the "Gemeinschaft zur Förderung von Pflanzeninnovation e.

CO N FLI C T O F I NTE R E S T
There is no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
Research data are not shared.