Plant virus infections control stomatal development

Stomata are important regulators of carbon dioxide uptake and transpirational water loss. They also represent points of vulnerability as bacterial and fungal pathogens utilise this natural opening as an entry portal, and thus have an increasingly complex relationship. Unlike the situation with bacterial and fungal pathogens, we know very little about the role of stomata in viral infection. Here we report findings showing that viral infection influences stomatal development in two susceptible host systems (Nicotiana tabacum with TMV (Tobacco mosaic virus), and Arabidopsis thaliana with TVCV (Turnip vein-clearing virus)), but not in resistant host systems (Nicotiana glutinosa and Chenopodium quinoa with TMV). Virus infected plants had significantly lower stomatal indices in systemic leaves of susceptible systems; N. tabacum 9.8% reduction and A. thaliana 12.3% reduction, but not in the resistant hosts. Stomatal density in systemic leaves was also significantly reduced in virus infected A. thaliana by 19.6% but not in N. tabacum or the resistant systems. In addition, transpiration rate was significantly reduced in TMV infected N. tabacum.

Scientific RepoRts | 6:34507 | DOI: 10.1038/srep34507 2 nd or 3 rd leaf to develop since inoculation; leaf impressions were taken of the abaxial surface and images were acquired from 3 areas of each leaf and SD (stomatal density = total number of stomata per mm 2 ) and SI (stomatal index = {(no. of stomata)/(no. of epidermal cells + no. of stomata)}*100) were calculated.
In contrast to the susceptible hosts, the resistant hosts N. glutinosa and C. quinoa did not exhibit a reduction in SI (N. glutinosa: n = 18, P = 0.581; Fig. 3a, C. quinoa: n = 17-19, P = 0.457; Fig. 3c) or SD (N. glutinosa: n = 18, P = 0.872; Fig. 3b, C. quinoa: n = 17-19, P = 0.215; Fig. 3d) in the systemic leaves when the plants were infected with TMV 7 dpi. This revealed that in marked contrast to the susceptible host there was no significant change in either SD or SI in the systemic leaves, which remain virus free (Fig. 3a-d). Transpiration assays. In order to investigate whether viral-induced changes in stomatal development impact on plant water relations transpiration rates were measured in N. tabacum healthy and TMV-infected plants in a separate experiment to those described above. TMV-infected N. tabacum plants exhibited significantly

Discussion
Stomatal development is influenced by a range of environmental cues and local signals 4,5,16 . Recently there has been considerable attention paid to the role of stomata during infection particularly by fungi and bacteria 15 . However much less attention has been devoted to the effects of viral infection on stomatal development. Previous investigations have briefly noted some viral induced developmental changes, in that Sweet potato feathery mottle virus (SPFMV) infected sweet potato plants have fewer, smaller, stomata 17 , while strawberry plants infected with a trio of viruses (Strawberry crinkle virus (SCrV), Strawberry mottle virus (SMV) and Strawberry mild yellow edge virus (SMYEV)) had fewer stomata 18 . In addition tobacco plants infected with a novel TMV strain from Egypt had fewer stomata 19  Ruggenthaler et al. 21 first demonstrated a possible molecular link between plant virus infections and stomatal development 21 . When the gene for the protein AtMBP2C (host protein shown to associate with the movement protein of TMV) was overexpressed in A. thaliana the mutants showed abnormal stomatal patterning. This was apparent with increased levels of paired stomata compared with WT, and low levels of triplet and quartette stomatal clusters were also found in the mutant 21 .
Here we extend these studies and present insights into the physiological responses during viral infection, likely as a consequence of altered stomatal development. We show that viral infection in two susceptible host types; A. thaliana and N. tabacum infected with TVCV and TMV respectively is associated with a reduction in stomatal density (Arabidopsis) and index (both systems) and that this is a systemic response. In marked contrast in the resistant hosts, N. glutinosa and C. quinoa infected with TMV, there was no significant change in stomatal density or index in virus free systemic leaves of mock and virus inoculated plants. We also showed that a reduction in transpiration was observed in virus infected N. tabacum that exhibited reduced stomatal index/density at 2 weeks post inoculation.
The question of whether the reduction in stomatal index seen in the infection on susceptible hosts is part of a defence response by the plant or is of benefit to the virus during infection will be the subject of future investigations. In the absence of other compensatory factors a reduction in stomatal density and index would likely be associated with reduced transpiration and possible increased host water use efficiency [10][11][12]22 .  A reduction in transpiration could also lead to reduced viral movement around the host. Viruses such as TMV are known to travel long distances through the phloem 23 . However, increasingly plant viruses are being found in the xylem and are capable of generating systemic infection via xylem transport 24,25 . In addition, transpiration rate will affect the movement of the virus through the apoplastic pathway and horizontal transportation from the phloem into the surrounding tissues 26 . A reduction in stomatal index could also be a response from the host to prevent further infection. Many pathogens use stomata as an entry portal to the plant host including bacteria, fungi, nematodes and protists 27 .
In summary, here we report that during viral infection, in susceptible hosts, there is a decrease in the number of stomata that develop on the leaf surface of leaves that have developed post-infection indicating that this is a systemic response.  Virus inoculation methods. Virus inoculum was prepared for manual inoculations by homogenising a small amount of infected leaf tissue (N. tabacum leaves for TMV, N. benthamiana leaves for TVCV) in sterile water. When plants were either 3-4 weeks old (A. thaliana) or 4-6 weeks old (Nicotiana spp), a healthy leaf was then chosen for inoculation, typically 2 nd or 3 rd newest leaf, which was then labelled by piercing a small hole using a sterile pipette tip, and dusted with carborundum powder. The virus homogenate was applied by gently stroking it onto the leaf with a gloved finger. After several minutes the inoculated leaf was washed with sterile water and the plant was left at specified growth conditions to allow development of infection. Mock inoculations received the same treatment but sterile water was used instead of virus inoculum.

Stomatal development analysis. Dental resin (President Jet Light Body, Coltène/Whaledent, Burgess
Hill, UK) was applied to the underside of the 3 rd or 4 th leaf that had developed since inoculation. In susceptible host species (N. tabacum and A. thaliana) this was 21 dpi and in resistant host species (N. glutinosa and C. quinoa) this was 7 dpi; this discrepancy was due to the nature of infections established in the respective hosts. In susceptible species, as the virus causes systemic infection and symptoms take 2-3 weeks to appear post inoculation, whereas the resistant host species produce symptoms after 1-3 dpi, as the asymptomatic leaves are virus free. Once set, the leaf was removed and colourless nail polish (Revlon) was coated on the impression. After drying, colourless sticky tape (Sellotape) was used to transfer the secondary impression onto a microscope slide and stuck down.
Slides were imaged using an inverted Zeiss Axiovert 200 M microscope driven by Volocity software (Improvision Ltd, Coventry, UK). For each slide, 3 areas were captured (top, middle and bottom). From these images, cell counts and size measurements could be made using Volocity Demo, and stomatal index (SI = {(no. of stomata)/(no. of epidermal cells + no. of stomata)}*100) and density (SD = total number of stomata per mm 2 ) could be calculated. Student's t-tests were used to asses significant differences (SI data required transformation and was subjected to arcsine square root transformation, formula used: = ASIN[SQRT(Stomatal Index/100)]) and data was tested for normality before applying the statistical test. Statistics were performed in Microsoft Excel (2010) and IBM SPSS Statistics 21.
For A. thaliana and N. tabacum plants, data was collated from 3 independent replicates into a mega-analysis. Resistant hosts N. glutinosa and C. quinoa contained compiled data from 2 biological replicates.
Stomatal size analysis was performed by measuring 15 stomata per individual plant (5 stomata per image, 3 areas captured per leaf) and the length of each stomata was recorded. Data in this section is from one biological replicate for each species. The data was tested for normality and a student's t-test was used to assess differences in means.
Transpiration assay. Transpiration assays were performed on N. tabacum plants that were mock inoculated or infected with TMV, 14 dpi. Duplicate sets of plants were either inoculated with TMV or mock inoculated with sterile water, after 14 days the pots were wrapped in plastic and sealed around the stem. Mass was recorded daily on a balance for 5 consecutive days. Measurements were recorded for each individual and regression plots were formed and homogeneity of slopes was measured using STATGAPHICS-Centurion. Data was compiled from 2 biological replicates.