Structural and Functional Responses of Soil Microbial Communities to Biodegradable Plastic Film Mulching in Two Agroecosystems

Polyethylene (PE) plastic mulch films are used globally in crop production but incur considerable disposal and environmental pollution issues. Biodegradable plastic mulch films (BDMs), an alternative to PE-based films, are designed to be tilled into the soil where they are expected to be mineralized to carbon dioxide, water and microbial biomass. However inadequate research regarding the impacts of repeated incorporation of BDMs on soil microbial communities has partly contributed to limited adoption of BDMs. In this study, we evaluated the effects of BDM incorporation on soil microbial community structure and function over two years in two geographical locations: Knoxville, TN, and in Mount Vernon, WA. Treatments included four plastic BDMs, a completely biodegradable cellulose mulch, a non-biodegradable PE mulch and a no mulch plot. Bacterial community structure determined using 16S rRNA amplicon sequencing revealed significant differences by location and season. Differences in bacterial communities by mulch treatment were not significant for any season in either location, except for Fall 2015 in WA where differences were observed between BDMs and no-mulch plots. Extracellular enzyme rate assays were used to characterize communities functionally, revealing significant differences by location and sampling season in both TN and WA but minimal differences between BDMs and PE treatments. Limited effects of BDM incorporation on soil bacterial community structure and soil enzyme activities when compared to PE suggest that BDMs have comparable influences on soil microbial communities, and therefore could be considered an alternative to PE. Importance Plastic film mulches increase crop yields and improve fruit quality. Most plastic mulches are made of polyethylene (PE), which is poorly degradable, resulting in undesirable end-of-life outcomes. Biodegradable mulches (BDMs) may be a sustainable alternative to PE. BDMs are made of polymers which can be degraded by soil microbial enzymes, and are meant to be tilled into soil after use. However, uncertainty about impacts of tilled-in BDMs on soil health has restricted adoption of BDMs. Our previous research showed BDMs did not have a major effect on a wide range of soil quality indicators (Sintim et al. 2019); here we focus on soil microbial communities, showing that BDMs do not have detectable effects on soil microbial communities and their functions, at least over the short term. This informs growers and regulators about use of BDMs in crop production, paving a way for an agricultural practice that reduces environmental plastic pollution.

regarding the impacts of repeated incorporation of BDMs on soil microbial communities has partly 23 contributed to limited adoption of BDMs. In this study, we evaluated the effects of BDM 24 incorporation on soil microbial community structure and function over two years in two 25 geographical locations: Knoxville, TN, and in Mount Vernon, WA. Treatments included four 26 plastic BDMs, a completely biodegradable cellulose mulch, a non-biodegradable PE mulch and a 27 no mulch plot. Bacterial community structure determined using 16S rRNA amplicon sequencing 28 revealed significant differences by location and season. Differences in bacterial communities by 29 mulch treatment were not significant for any season in either location, except for Fall 2015 in WA 30 where differences were observed between BDMs and no-mulch plots. Extracellular enzyme rate 31 assays were used to characterize communities functionally, revealing significant differences by 32 location and sampling season in both TN and WA but minimal differences between BDMs and PE 33 treatments. Limited effects of BDM incorporation on soil bacterial community structure and soil 34 enzyme activities when compared to PE suggest that BDMs have comparable influences on soil 35 microbial communities, and therefore could be considered an alternative to PE. 36 Importance 37 Plastic film mulches increase crop yields and improve fruit quality. Most plastic mulches are made 38 of polyethylene (PE), which is poorly degradable, resulting in undesirable end-of-life outcomes. 39 Biodegradable mulches (BDMs) may be a sustainable alternative to PE. BDMs are made of The mulches were tested in the field over two years (2015 to 2016) under pie pumpkin (Cucurbita 109 pepo) as a test crop, with full experimental details described in Sintim et al. (31) and S. Ghimire Fluvaquents. Henceforth in the paper, Knoxville will be referred to as TN and Mount Vernon will 117 be referred to as WA. 118 Each field site was arranged as a randomized complete block design with four replications of seven 119 main plot treatments (six mulch treatments and one no mulch control). Before mulch application 120 began in TN and WA, the plots were under winter wheat (Triticum aestivum) cover crop in TN 121 and clover (Trifolium spp.) at WA. Mulches were machine-laid on raised beds. Pumpkins 122 (Cucurbita pepo) were grown during the growing season. The PE mulch was removed after 123 pumpkin harvest, while the BDMs were tilled into the soil with a rototiller. 124 Soil water content and temperature were monitored as described in Sintim et al. (31). Briefly, 125 sensors (5TM, Decagon Devices Inc., Pullman, WA) installed in the center of each mulch 126 treatment at 10-cm and 20-cm soil depths for one field block were connected to data loggers 2.5 Quantitative PCR for bacterial and fungal abundances 154 As a proxy for bacterial and fungal abundances, 16S rRNA (bacteria) and ITS (fungi) gene copy 155 abundances were quantified from soil DNA samples using Femto™ Bacterial DNA quantification 156 kit (Zymo Research) and Femto™ Fungal DNA quantification kit (Zymo Research) following the 157 manufacturer's protocol. DNA extracts were diluted 1:10 prior to quantification and 1 µl of the 158 diluted samples was used for each qPCR reaction. All samples were analyzed in triplicate. standards and substrates were added to the soil slurries. The plates were sealed and inverted to mix 201 the contents. Incubation was done for 3 hours at room temperature, after which the substrate and 202 standard plates were centrifuged at 1500 rpm (~327 x g) for 3 min. The supernatants were pipetted   was not included in this model as PERMANOVA results from PRIMER-E were used to report differences between locations. Boxplots for equal variance and outliers, reported in SAS, were 244 used to remove outliers in the dataset. Normality was checked using Shapiro-Wilk test (W > 0.9) 245 and probability plots for residuals, and equal variance using Levene's test (α = 0.05). Data were 246 log transformed as necessary when these conditions were not met. Raw experimental values and 247 standard errors are reported in the figures. All graphics were plotted using R. v. 3.4.0. Type III 248 tests of fixed effects and interaction effects are reported.

249
To assess for potential enrichment of bacteria and fungi, a paired t-test was conducted using initial

Environmental and soil physicochemical data 259
Environmental data collected during the experiment is reported in Sintim et al. (31) and in Table   260 S1. The mean daily air temperature in Knoxville, TN during experimental years of 2015 to 2016 261 was about 4 ºC higher than in Mount Vernon, WA (Table S1). The total annual precipitation during 262 the experimental years was higher in Knoxville, TN than in Mount Vernon, WA.

263
Soil temperature, moisture and physicochemical properties were measured and reported previously 264 by Sintim et al. (31). In summary, significantly increased soil temperature was observed in the early growing seasons in the plastic mulch plots compared to the cellulose and no-mulch plots. On 266 average, the monthly soil temperature was greater in TN than in WA. The soil water content varied 267 more among the mulch treatments, with PE mulch having the highest soil water content for the 268 greatest period of time. Mulched plots generally had higher water content than the no mulch plots.

269
The soil health analysis revealed some effects of mulching on certain properties (namely aggregate 270 stability, infiltration, soil pH, electrical conductivity, nitrate, and exchangeable potassium), but 271 these were not consistent among BDMs, nor across sampling times and locations. The NMDS ordination revealed a clear difference in community structure between TN and WA 274 when combining data from all four sampling seasons (Spring 2015 to Fall 2016) (Fig 1a). 275 Permutational ANOVA (PERMANOVA) tests confirmed significant differences between TN and 276 WA soil microbial communities (Table 3, Table S2). The mean relative abundances of the most the most influential OTUs contributing to the variation seen between location (Fig 1b). The most 279 influential OTUs belonged to several classes of microbes such as Acidobacteria_Gp7, 280 Acidobacteria_Gp16, Acidobacteria_Gp4, Planctomycetacia and Spartobacteria. CAP analysis 281 revealed that the differences in soil communities between TN and WA were most related to pH, 282 soil moisture and organic matter content: the communities in TN were related to increased pH, 283 whereas moisture and organic matter were positively related to communities in WA ( Fig S1).

284
In addition to locational differences, bacterial communities also differed significantly between the 285 different seasons (Table 3, Table S2). For both locations, Spring communities were more similar 286 to each other than Fall communities (Fig 2a, b). SIMPER tests revealed that several genera of were cumulatively responsible for 60% of the seasonal variance in bacterial communities ( Fig S2   289 and S3). Interestingly, Streptomyces spp. increased in percent relative abundance over time from 290 Spring 2015 to Fall 2016 in both TN and WA (Fig S2).

291
Unlike location and season, the mulch treatments did not have a significant effect on bacterial 292 community structure (Fig 2). Because of the locational and seasonal differences, we additionally 293 analyzed each time-location set separately, and did not detect any significant effects of treatment 294 on community structure ( Fig S4, Table 3, Table S2).

295
Alpha diversity of the soil bacterial communities was estimated using observed species richness 296 and inverse Simpson index of diversity (Table S3). The observed species richness estimator 297 measures count of unique OTUs in each sample. There were significant differences between TN 298 and WA (p < 0.05) in richness estimates (Table 4, Fig 3a). TN had greater richness than WA 299 throughout the experiment, ranging from 260 to 300 unique OTUs. WA richness estimates ranged 300 from 250 to 280 OTUs over the two years. The locational differences in richness were due to a 301 lower richness in Fall 2015, Spring 2016 and Fall 2016 in WA (Fig 3). The Inverse Simpson 302 diversity index ranges were similar between TN and WA, ranging from 7 to 11.

303
For both TN and WA, there was a significant difference between seasons in both richness and 304 inverse Simpson index (Table 4). The richness estimates in TN significantly differed between 2015 305 and 2016 (Fig 3a). In WA, Fall 2015 differed in richness from the other time points. In TN, Fall 306 2016 diversity was significantly higher than other seasons. Diversity estimates were significantly 307 lower in Spring than in the Fall seasons for WA (Fig 3b).

308
In TN, PE had the lowest richness and BioAgri had the highest, however, treatment differences in 309 richness estimates were not significant (Table 4) when analyzing data using a mixed model.

310
Inverse Simpson diversity indices were also not significantly different between treatments (Table   311   4). Looking at the final time point in TN, diversity estimates were highest for Weedguard, and 312 lowest for PE, and in WA, the estimates were highest for Weedguard, followed by PE with BDMs 313 having lower diversity than PE or Weedguard, however these differences were not significant (Fig   314   3b). for each mulch treatment, a paired t-test was used to identify differences significantly different 320 from zero (Table 5). There was a significant increase in bacterial gene copies under BDM and 321 Weedguard treatments in WA, but no significant change for no mulch and PE treatments (Table   322   5). There was also a significant enrichment in fungal gene copies over time for two of the BDMs 323 (PLA/PHA and Naturecycle) in WA. In TN, significant enrichment in bacterial gene copies was 324 seen under Organix, PLA+PLA and PE treatments (Table 5) but no enrichment was seen in fungal 325 gene copies. In order to determine if these changes were significantly different between treatments, 326 the differences between the final (Fall 2016) and the initial (Spring 2015) abundances were 327 analyzed using a mixed model analysis of variance in SAS v 9.3 and Tukey post hoc tests. In both 328 locations, mulch treatments did not have a significant effect on the changes in either 16S or ITS 329 gene copies over the course of the experiment (Fig 4a, b).

4 Microbial community functions
To assess potential functional responses of the soil microbial communities, extracellular enzyme 332 potential rate assays were conducted for common carbon, nitrogen, and phosphorus cycling 333 enzymes in soil ( Table 2). The data were combined over the two years to visualize Bray Curtis 334 similarities of the enzyme rate profiles (Fig 5). Locational differences in the enzyme profile were 335 significant (p < 0.05), as were seasonal differences in both TN (p < 0.05) and WA (p < 0.05) 336 evaluated using PERMANOVA (Fig 5). However, mulch treatment did not have a significant

340
In general, the enzyme activity rates oscillated between higher activities in the Spring and lower 341 activities in the Fall. When analyzed separately for each enzyme, the data over the two years 342 revealed a significant effect of sampling time in TN for all seven enzymes assayed. In WA, enzyme 343 activities of β-xylosidase, β-glucosidase, α-glucosidase, N-acetyl β glucosaminidase and 344 phosphatase were significantly different between sampling times (Fig 6). In WA, cellobiosidase 345 and leucine amino peptidase activities remained unchanged across the seasons (10-22 nmol 346 activity g -1 dry soil h -1 for cellobiosidase and 200-375 nmol activity g -1 dry soil h -1 for leucine 347 amino peptidase) (Fig 6).

348
When averaged across seasons, mulch treatment differences were not significant for any soil 349 enzymes in WA (Table 6). However, in TN, an effect of mulch treatment was observed for N-350 acetyl β glucosaminidase activities (Table 6). N-acetyl β glucosaminidase activity was reduced 351 under BDMs and PE compared to no mulch plots. Interaction effects of mulch treatment and time 352 of sampling were not detectable for any of the enzymes assayed in TN or WA (Table 6). soil burial studies with an artificially high 1.8 g PBAT films in 300 g soil with added 30 ml of 367 basal medium. Finally, our aim was to characterize responses in bulk soil communities to 368 understand the overall system level response to plastic films, so we likely missed changes 369 happening on smaller spatial scales. For example, Li et al. (29) reported changes in microbial 370 communities in soils that were sampled in close proximity to buried mulch films, indicating that 371 microbial communities in the immediate vicinity of the films may be affected. Here we show that 372 any local effects of mulch films are not detectable at a plot/field scale, at least over a 2-year period. 373 We did note significant differences in soil bacterial composition by location and season, which has 374 been observed in other studies (29, 30), confirming that mulch effects are minimal compared to 375 other drivers of community structure variation. It is well accepted that local soil conditions such 376 as temperature, moisture and pH play a pivotal role in shaping microbial communities (30,44,45).
In this study, the location differences in communities were attributed to higher relative abundances 378 of Acidobacteria, Actinobacteria and Planctomycetes in TN and higher abundances of β-and γ-379 Proteobacteria in WA. This corresponds with higher pH and saturated K in TN and higher soil  However, because we did not observe differences in the relative abundance of this taxa between 393 BDMs, PE or no mulch control, this increase is likely attributable to the agronomic management 394 of the plots (e.g. plant species, irrigation or fertilizer regimes etc.).

395
Richness estimates showed significant differences across locations and seasons in TN and WA.  There is also precedent for the differential responses in microbial enrichment we observed between 422 the two locations, with both fungal and bacterial enrichment in WA, but only bacterial enrichment in TN. In a similar study comparing BDM effects in three locations, it was found that BDMs 424 resulted in soil fungal enrichment in Texas and bacterial enrichment in TN (29). In one study, soil 425 pH was shown to be the best predictor of bacterial community composition across different land 426 use types, while fungal communities were shown to be most closely associated with changes in 427 soil nutrient status such as extractable P concentrations and C:N ratios (53). Both TN and WA 428 soils had comparable fungal gene abundances initially in the Spring of 2015. However, since the 429 microbial communities in WA were seen to be controlled by the presence of organic matter (Fig   430   S1) and WA soils had higher C:N ratios than TN soils this could have contributed to a fungal 431 enrichment in WA but not in TN.

432
Enzyme assays were conducted to assess potential activity rates for common carbon, nitrogen and 433 phosphorus cycling enzymes in soil. As with bacterial community structure, enzyme activity 434 profiles showed the greatest differences by location and season (Fig 5, Table 6). The seasonal 435 oscillation in enzyme activities seen for almost all the enzymes could be attributed to seasonal 436 tillage operations which tend to offset many of the soil biological functions (54-56) (Fig 6). This higher temperatures lower growth rates are found (60). This decrease in growth rate was shown to 456 be more drastic for fungi than for bacteria, resulting in an increase in the ratio of bacterial to fungal 457 growth rate at higher temperatures. Thus, the high temperatures under mulches in the summer in 458 TN were above optimum growth conditions for soil microbes and may have reduced soil enzyme 459 activities. Cold-adapted microorganisms, which are expected to be more prevalent at the WA site, 460 tend to respond more efficiently to increased temperature than warm-adapted microbes (61). The 461 greatest relative temperature sensitivity of decomposition processes has been observed at low 462 temperatures (62). Warming experiments have revealed reduced xylosidase activity in soils (5-15 463 cm deep) under medium-warmed plots compared to unwarmed plots (59). It has also been reported 464 that warming induces decreases in the temperature sensitivity of β-xylosidase activity in the H 465 horizon (63). One study reported greater increase of the relative temperature sensitivity of XYL 466 and NAG (important for C cycling) at lower temperatures, compared to amino peptidase enzymes 467 suggesting that temperature plays a pivotal role in regulating the use of substrates. Thus, the turnover of easily degradable C substrates (like glucose) is more sensitive to temperature than 469 higher molecular compounds, at least for cold soils (64).

470
Looking specifically at studies which assessed soil enzyme activities after treatment with 471 biodegradable plastic film, one field study reported that soil microbial biomass and beta-472 glucosidase activity were most responsive to mulch; however that study did not have PE as a 473 control, so it is unclear if this response was specific to BDMs or just related to plastic mulching 474 generally (65). That study also focused on soils in close proximity to plastic, rather than bulk soil

489
Two years of biodegradable and PE mulch treatments in a vegetable agroecosystem in two 490 locations revealed some minor effects on soil microbial communities and their functions. While we were not able to detect any significant effect of plastic mulches on bacterial community 492 structure, richness or diversity, we did observe other impacts on the communities that were 493 location-dependent. In particular, we noted that in WA, biodegradable mulches enriched for both 494 bacteria and fungi, suggesting a response to BDM incorporation into soils; in contrast only 495 bacterial enrichment was apparent in TN, and only for three of the five plastics tested. We             Table   639 6). Error bars indicate SEM of four replicate samples.