Preharvest Farming Practices Impacting Fresh Produce Safety

ABSTRACT

Advancements in agriculture and food processing techniques have been instrumental in the development of modern human societies. Vast improvements in agronomic practices, handling, and processing have allowed us to produce and preserve mass quantities of food. Yet despite all these innovations and potentially as a consequence of these mass production practices, more and more outbreaks of human pathogens linked to raw and processed foods are identified every year. It is evident that our increased capacity for microbial detection has contributed to the greater number of outbreaks detected. However, our understanding of how these events originate and what agronomic, packaging, and environmental factors influence the survival, persistence, and proliferation of human pathogens remains of scientific debate. This review seeks to identify those past and current challenges to the safety of fresh produce and focuses on production practices and how those impact produce safety. It reflects on 20 years of research, industry guidelines, and federal standards and how they have evolved to our current understanding of fresh produce safety. This document is not intended to summarize and describe all fruit and vegetable farming practices across the United States and the rest of the world. We understand the significant differences in production practices that exist across regions. This review highlights those general farming practices that significantly impact past and current food safety issues. It focuses on current and future research needs and on preharvest food safety control measures in fresh-produce safety that could provide insight into the mechanisms of pathogen contamination, survival, and inactivation under field and packinghouse conditions.

OVERVIEW OF PREHARVEST FOOD SAFETY

Foodborne illness associated with fresh fruits and vegetables and growing and harvesting practices that impact the microbial safety of produce have been under the scrutiny of federal and state regulators for over 40 years. In 1998 the Food and Drug Administration published the Guide to Minimize Microbial Food Safety Hazards for Fruits and Vegetables (GMMFSH) in response to the U.S. President’s Food Safety Initiative, which launched, among other programs, the “Fight Bac” campaign. This GMMFSH guideline was created based on outbreak investigations that identified fresh produce as “an area of concern” (1) and summarized the then-current body of research and understanding of the risk factors associated with agricultural practices and how they could impact contamination of fresh produce with human pathogens. At the time, the GMMFSH guidelines did not address practices for risk elimination, supply chain, or environmental contaminants; however, they clearly identified the foundation for future research and new federal policies (the Food Safety Modernization Act [FSMA] and the Produce Safety Rule signed into law in 2011, which are currently impacting the farm to fork continuum). Twenty years after the development of these guidelines it is evident that previously identified sources of contamination linked to agronomic practices continue to be implicated in many of the more than 140 produce-related outbreaks investigated by federal and state agencies (2).

During these 2 decades, there has been an increase in produce contamination events that could be attributed to any of the following factors: (i) improved detection platforms, (ii) increased consumption of fresh produce and fresh-cut products, (iii) transport and distribution of fresh produce from many regions around the world, and (iv) production and potentially harvest and packing practices that either did not present any past problems or that continue to present issues due to continuing lack of understanding of the factors that impact contamination despite industry-driven practices that have improved food safety systems (good agricultural practices, hazard analysis of critical control points, and other third-party food safety schemes).

Better surveillance and detection systems have significantly impacted our ability to identify different microbial contaminants and impacted the increase in the number of pathogen contamination events detected each year. A good example that highlights our increased ability to detect different pathogens is reflected in the number of Cyclospora outbreaks detected between 2013 and 2018 and Listeria monocytogenes between 2011 and 2018 (3) because of the use of modern technologies to generate genome sequences, identify pathogenic mechanisms, and provide evidence-based risk assessments. This increase in detection capacity has not translated into an increased ability to identify sources of contamination or risk factors impacting pathogen outbreaks or into reducing trends in foodborne diseases. Table 1 summarizes areas of concern within soil, plant, water, and packinghouse environments that impact survival, internalization, and interactions of human pathogens in different anthropogenic environments.

TABLE 1.

Challenges within the fresh produce continuum

Environment or Condition	Challenge areas
Soil	Soil properties, biological soil amendments, compost, raw manure, herbal teas, and persistence of human pathogens vs. crop contamination
Farm history	Farms formerly used for the rearing of animals or those in close proximity to small, medium, and large CAFOsWild and domestic animal field intrusion and field contamination
Crop	Growth and survival of human pathogens on crop inputs, including organic and chemical fertilizers and pesticidesGrowing practices and whether organic, conventional, biodynamic, sustainable, or other agronomic practices impact pathogen contamination, survival, persistence, and distribution within the farm
Water	Microbial quality based on source, region, activity, and practice and the equipment used to deliver water to the crop (pumps, drip tape, pipeline, emitters, sand and plate filters, and injectors of sanitizers
Worker	Health and hygiene, especially in areas where field and harvest crews move from state to state or between countries while receiving multiple food safety trainings in different languages that in many cases cause fatigue and confusion between growing and harvesting practices
Packinghouse	Harvest equipment and containersPackinghouse equipment, sanitation, and pest controlProper use of sanitizers and understanding the mechanism and best practices to reduce cross-contamination within sanitation and washing systems
Climate change and globalization	Population growth and demographics, especially in countries with aging populations and with immune-compromised conditionsGlobalization, which allows continuing movement of individuals carrying multiple microorganisms across the worldChanging eating habits (consumption of raw or lightly cooked food)Climate change, water loss, and continuing loss of wildlife habitat converted into farmland that increases the impact of pest control able to support large-scale farming practices

This review is not intended to provide a detailed description of grower production practices and all the potential interactions that impact fresh produce safety. It should not be used as a guide for safely growing fruits and vegetables. Instead, this review largely focuses on those farming practices and research components needed to elucidate pathogen interactions at the farm level, summarizes some of the most important past and current research topics, and provides additional information based on observation and experience of the sources of contamination associated with different farming practices. We advise the reader to also review other publications (4–13) to gain a broader perspective of the issues addressed here.

Fruits and Vegetables in the United States Impacted by Microbial Contamination

According to the U.S. 2012 Census of Agriculture, fruit, tree nut, and berry sales together amounted to $25.9 billion, while vegetable sales were estimated at $16.9 billion. Ten states account for 94% of all fruit, tree nut, and berry sales, with California, Washington, and Florida controlling 87% of these sales. Oranges and grapefruits are by far the largest citrus crops. Among non-citrus fruits, berries, grapes, apples, peaches, sweet cherries, blueberries, avocados, and strawberries are the largest commodities. Almonds, pecans, walnuts, and pistachios account for almost all the tree nut acreage in the United States. The top 10 vegetable-producing states accounted for 77% of sales in 2012, with California alone accounting for 38% of total sales, followed by Florida and Washington. The major vegetable crops are potatoes, sweet corn, tomatoes, lettuce, and watermelon (14).

Fresh produce represents one of the major causes of foodborne illnesses (46%) in the United States (15). The U.S. Centers for Disease Control and Prevention (CDC) estimates that each year 48 million Americans get foodborne diseases, resulting in 128,000 hospitalizations and 3,000 deaths. According to the U.S. Food and Drug Administration (FDA), from 1996 to 2010, produce was related to approximately 131 reported outbreaks, causing 1,382 hospitalizations and 34 deaths. In 2015, approximately 37 outbreaks resulting in 1,800 illnesses were associated with fresh produce. The most outbreak-associated illness was from seeded vegetables (e.g., cucumbers and tomatoes; 1,121 illnesses) and vegetable row crops (e.g., leafy vegetables; 383 illnesses) (3). Leafy vegetables were responsible for 22% of all foodborne illness in the United States and were the second most frequent cause of hospitalizations (14%) between 1998 and 2008 (15). In general, the fresh produce most frequently associated with foodborne illnesses included leafy greens, cantaloupes, tomatoes, green onions, and herbs, accounting for more than 80% of all produce-associated outbreaks (16, 17). Despite the significant number of illnesses detected each year, a larger and significant number of foodborne diseases go unreported because of the process of reporting and identifying the disease, where health care providers need to test stool samples for the right pathogen and then report the infection to local health departments. This process accounts for about 20,000 cases specific to food poisoning that are properly diagnosed, while the majority of cases remain unreported or undiagnosed (2, 3, 15, 18).

Foodborne pathogens commonly implicated in the majority of the outbreaks include Shiga toxin-producing Escherichia coli, Salmonella, and L. monocytogenes. Preharvest contamination of produce with foodborne pathogens has been a major food safety issue. Potential sources for foodborne pathogens are fecal materials and soil (19–21), irrigation water (20, 22, 23), improperly composted manure (20–22), air, wild and domestic animals, birds, rodents and flies, equipment, and human handling (24, 25). Several factors such as seasonal and environmental conditions, temperature, and type of produce may influence the prevalence of pathogens at preharvest stages (26, 27). Since fresh produce is consumed raw, understanding the potential sources of pathogens during growing, harvesting, and packing is vital to develop risk management strategies.

FSMA was developed following the scientific evidence available as of 2011, when it was signed into law. Within these regulations, guidelines specific to the microbial quality of water; sampling size and testing methods; the prevalence and persistence of pathogens in soil; management practices and applications of raw, aged, and composted manure; soil remediation; and sanitation practices may not provide sufficient hurdles to prevent contamination. Consequently, we continue to need the development of funding modules for researchers that can perform multidisciplinary research on industry practices to provide pathogen reduction solutions at the farm level.

PREHARVEST FOOD SAFETY: POTENTIAL ROUTES OF MICROBIAL CONTAMINATION

The diversity and scale of farming operations at local, regional, and international locations coupled with various environmental factors and plant, irrigation water, and soil characteristics provide a significant obstacle to listing, describing, and evaluating the complexity of interactions and their impacts on enteric human pathogen risk assessment and management in fresh produce. Such compounding conditions also impact the implementation, development, and adequate assessment of government regulations that look to standardize food safety guidelines across the globe.

It is well recognized that fresh produce-related foodborne illnesses in the United States are associated with microbial pathogens of animal origin, while imported produce tends to be associated with human contamination (28). These potential differences present two questions. The first is, What is the impact of growing practices on actual contamination and transmission of human pathogens to produce on imported food? The second is, Irrespective of whether produce is imported or produced domestically, if both have similar growing and packaging processes; can we infer that pathogen contamination is mainly associated with growing conditions and length in transport instead of human intervention? In both systems, contamination could be occurring from growing or packaging practices and only becoming significant to consumers once extended transport and length of storage, mixed with consumer manipulation, provide the necessary conditions to cause human disease. Based on these potential outcomes, what we could be missing is a clear understanding of how human pathogens, once in produce (field contamination), interact with packing environments, temperature, and phyllosphere or rhizosphere communities and which of these factors masks and/or prepares enteric pathogens to persist during storage and transport.

The presence of a variety of nutrient-rich exudates in the phyllosphere, primarily composed of sugars and organic and amino acids, are known to impact well-established microbial communities and potentially human pathogens. It has been suggested that microbial populations in plants could be modified by directly changing the nutrient availability on the leaf surface. For epiphytic and plant pathogenic bacteria, it is recognized that changes in leaf nutritional composition, particularly the abundance of carbon sources (sugars), coupled with alterations in leaf morphology, significantly influence the capacity of bacteria to colonize leaves and determine their population size and spatial distribution. Fertility management, particularly nitrogen levels and forms, influences the abundance of these sugars and the size and shape of intercellular spacing, water congestion, and subsequent bacterial colonization, including human pathogens (73). For epiphytes and human pathogens to gain access to these nutrient-rich islands after harvesting, handling, and packing, the proximity and time of exposure to these nutrient oases, in combination with moisture content, impact the balance toward increased persistence. Current scientific evidence has identified different sources of contamination; however, the specific influence and interaction among production environments and crop management practices continues to be misunderstood and subject to scientific debate.

One of the least-studied issues concerning fresh produce safety is related to stress conditions (sanitizers, desiccation and low temperatures, frost protection, high UV index, and long-distance transport), which injures cells and triggers the viable but nonculturable (VBNC) state, and human pathogen fitness. Over 35 human pathogens are known to enter the VBNC state under stress conditions in produce (29), but the impact on human pathogen outbreaks is unknown. Dinu and Bach (30) identified different cells of E. coli O157:H7 induced into the VBNC state by exposing them to low temperature on the surface of lettuce and spinach plants. Masmoudi et al. (31), Dinu and Bach (30), Gião and Keevil (32), and Liu et al. (33) found similar conditions that would trigger the VBNC state in L. monocytogenes and Staphylococcus aureus, especially under low temperatures, suggesting that similar environmental conditions, including frost protection events, dry and wet conditions, transport, and storage, could also trigger VBNC under open field environments.

There is no clear evidence to prove that resuscitation of foodborne pathogens is directly associated with human diseases, but many pathogens retain virulence in the VBNC state (34, 35), and some may have been implicated in foodborne illnesses, as suggested by Makino et al. (36) with E. coli O157:H7 and Salmonella and with E. coli O104:H4. Infectious doses for pathogens such as E. coli O157:H7, Salmonella, and L. monocytogenes range in the 10 to 1,000 cell threshold; however, it is not known if there is a similar threshold for VBNC cells from these pathogens and whether they can directly cause foodborne outbreaks. Further research is needed to understand the pathogenicity of VBNC cells, their infectious dose, and their interaction with farming practices, including their survival in water, soil, biological soil amendments, and packinghouse environments.

Water

Food security and safety are intertwined with water, one of our most valuable natural resources. Our ability to provide reliable access to sufficient, affordable, safe, and good-quality food is determined mainly by land and water availability (37). Water, either from rain or commercial methods for sourcing, using, and applying this natural resource, is essential for food security at a global scale. Water withdrawal and consumption is dominated by agriculture, which controls between 70 and 90% of its use (38). Key components of our ability to use water in agriculture include source, volume, and microbial quality to prevent the contamination of food with human pathogens. In the United States, the current standards for agricultural water established by the FSMA Produce Safety Rule identify generic E. coli as the indicator of choice to predict water contamination with any human pathogen. Current standards for generic E. coli establish the following numerical criterion: geometric mean of ≤126 CFU/100 ml of water and a statistical threshold value of ≤410 CFU/100 ml of water. Both of these values need to be met irrespective of the water source used for irrigation if this water is to be applied to the harvestable portion of a crop (39). These standards were established following the Environmental Protection Agency’s recreational water criteria, which take into consideration epidemiological studies and current and past scientific evidence describing how different individuals become ill by swallowing recreational water that is contaminated with feces and by direct human body contact with this water (18, 39). Despite these important efforts to standardize the microbial quality of water, it is clear that these standards were created based on recreational water parameters, with little information on how this microbiological criterion affects pathogen concentrations, survival, and persistence on produce and what impact it may have on consumer safety. National and international farmers will need to comply with these standards in the near future, but if this is not immediately possible, the FDA established three alternative provisions that growers can follow while they address the nature of contamination and develop corrective actions as soon as is practicable, but no later than the following year (39). Farmers using water that does not meet these microbiological criteria can (i) allow a maximum 4-day interval between the last irrigation and the harvest of the crop to allow potentially dangerous microorganisms to die off in the field, (ii) allow time between harvest and the end of storage for potentially dangerous microorganisms to die off or to be removed during commercial washing activities within appropriate limits, or (iii) treat the water. Most of these alternative provisions have not been tested by the scientific community, and it is unknown whether they are safe practices for growers and consumers.

When comparing the FSMA microbial water quality standards to WHO or Codex Alimentarius guidelines, there are major differences in the approach to defining “low-risk irrigation water.” However, they all seem to agree that the best practice is to evaluate the risk of contamination before using irrigation water that will be in direct contact with fresh fruits and vegetables. Since 1989, the WHO approach has used fecal coliforms and intestinal nematodes as the indicators of choice, with numerical standards following a geometric mean of ≤1,000 CFU/100 ml of water and ≤1 CFU/1,000 ml of water, respectively. WHO revised these standards in 2015, and their current understanding and guidelines no longer define a specific microbiological criterion. Instead, they point to the use of a risk analysis approach specific to each country to determine the actual risk of contamination of water used for irrigation in primary production (37, 40). In many regions of the world, this approach is unknown, and the microbiological criterion varies by country and in some instances are unknown.

The Codex Alimentarius has addressed the issue of microbial water quality in primary production in a different manner, providing only general guidelines to risk assessment instead of providing a specific numerical value. In these standards, the only defined numerical value refers to the use of potable/clean water when it will be in direct contact with the edible portion of fruits and vegetables (41, 42), regardless of source and method of application. Irrespective of the numerical criteria used to assess the microbial quality of irrigation water, correlations between the presence of generic E. coli at those numerical standards and the presence of other human pathogens in the water column or in water sediments coming from wells, rivers, or ponds used for irrigating produce is not significant (43–46) and fluctuates based on time of day, depth, location, rain, water flow, die-off rates of E. coli, and the overall diurnal distribution of this microorganism in the water column and sediments (47).

Die-off rates of generic E. coli and other enteric microorganisms have been shown to have a biphasic nature. The exponential nature of the reduction of these microorganisms under different environmental conditions suggests that relying on microbial die-off rates alone as a mitigation practice to reduce pathogen contamination could be problematic. Further, sample size, test method, location within the water reservoir impacting nutrient content in the sample (48), and growing season compound even more the potential determination and description of inactivation of generic E. coli and other enteric pathogens (12). These factors are known to impact how different microorganisms enter the VBNC state on different matrixes and whether their survival and virulence are enhanced or maintained under this cellular condition (29, 32).

The microbial quality of water sources (wells, rivers, and ponds) has also been shown to vary significantly, mainly from environmental inputs and intrusions, especially in rivers and ponds due to the proximity to wildlife, other animal inputs, runoff, and adjacent farming activities (49, 50). It is well established that surface waters (39, 51, 52) pose the biggest challenge for fruit and vegetable production because of numerous routes of contamination that producers cannot control. Rainfall events tend to increase or reduce the concentration of indicator and pathogenic microorganisms in the water column due to water runoff, sediment disturbance, dilution, and other inputs such as wastewater influents (9, 53, 54). Which factors increase or decrease concentrations seems to vary based on region and water source, and no clear associations can be used to describe overall patterns across regions and irrigation practices. In the case of well water, the main issues associated with enteric pathogen contamination are attributed to soil type, height of the water column, proximity to animal operations, and the design and construction of the well. Deeper wells tend to be less contaminated than shallow wells due to greater microbial infiltration through the soil profile (9).

Frost protection

In the southeastern United States and other global growing regions, the use of frost protection to reduce crop losses during freezing temperatures is a common practice (55). This treatment uses massive quantities of water, in some instances, for days or weeks depending on the weather pattern, with water demands in the rate of 60 gallons/acre/minute or more, depending on the size of the area that needs to be protected. To maintain such levels of water requirements, growers depend on surface water sources to protect their crops. Frost events can occur during blossom, fruit development, or close to harvest, and this method is typically used for strawberries, blueberries, apples, cherries, and peaches, among other crops (56). To date, there seems to be little to no scientific literature describing the impacts of these events on the overall safety of these crops. For example, it is not known whether enteric pathogens are present in those water sources during these cold weather events, and, if present, whether they are able to persist during single or multiple frost events. If present, do these enteric pathogens remain viable until harvest, and if they do, could they transfer to mechanical harvesting equipment, human hands (harvest crew), harvesting and picking baskets, or the packinghouse? These questions are of special importance with certain fruits such as fresh strawberries and blueberries, where further downstream, no commercial washing practices are available that could reduce the presence of enteric pathogens.

It is known that most of the human pathogen outbreaks associated with berries from 1983 to 2013 were linked to frozen rather than fresh berries. The majority of outbreaks have an unknown source of contamination; however, hands and field and processing practices have been suggested as the most common routes of contamination. Typically, consumers of fresh fruits and vegetables, including berries, gain access in supermarkets and other retail establishments to U.S. no. 1 produce, while any other grades tend to be used for processing and frozen product manufacturing. This does not mean that the safety and quality of the product is compromised, but further handling, selection, and in some instances, fruit washing (mainly for blueberries and strawberries) is performed before freezing. It is not known whether these outbreaks related to berries can be associated with field or processing contamination and whether frost protection was used or impacted the safety of the frozen product.

Fresh fruits, including peaches, apples and cherries, also receive further washing during postharvest handling and storage before they are sent to market. Under such conditions, deficiencies in packinghouse sanitation and equipment conditions could provide reservoirs and sources of enteric pathogen contamination, as seen in the 2014 caramel apple or stone fruit Listeriosis outbreaks (57, 58). To date, the origin of the initial contamination event in those two outbreaks is unknown, and further risk assessment and epidemiological studies are needed to identify those sources of contamination.

Overhead cooling and crop protection sprays

Water sources used for field overhead cooling and crop protection sprays mainly determine the potential route and source of contamination of fresh fruits and vegetables with enteric pathogens. In general, overhead cooling is used in orchards to reduce sunburn of exposed fruit and/or to enhance color development on certain fruits (59, 60), while crop sprays encompass applications of pesticide, growth regulators, bioactive natural compounds, or other chemicals that are used to modulate crop diseases and growth. Regardless of the final goal of the application method, the microbial quality of water and the developmental stage of the crop impact enteric pathogen contamination. As discussed previously, well and surface water sources pose different risk levels based on multiple environmental and anthropogenic inputs. Current harmonized good agricultural practices and the Produce Safety Rule have different microbial water quality standards when water will be in direct contact with the edible portion of the crop. In the case of harmonized good agricultural practices, this water, if used for cooling or crop sprays close to harvest, must be “microbial safe water,” which for growers and auditors tends to be represented by the use of potable water in these applications. However, the Produce Safety Rule provides a different microbiological water standard for water that will be used under the same conditions. This discrepancy is important since it has been shown that fungicides and other pesticides have little to no negative impact on the survival and persistence of human pathogens in these solutions (61, 62). Further, there is little information on the persistence of enteric pathogens in fruits or vegetables if present in and applied through pesticide solutions. Recent work by Lopez-Velasco et al. (63) showed that Salmonella applied to field-grown tomatoes through pesticide application was able to survive up to 15 days postinoculation and that 80% of the samples remained contaminated with this human pathogen even after postharvest washing with a sodium hypochlorite solution. These findings are remarkable when looking at the very low levels of Salmonella (log 2 CFU per fruit) inoculated onto the surface of tomatoes and the low levels of generic E. coli present in the system, suggesting that this enteric pathogen may be using these pesticides as nutrient sources, allowing it to potentially remain viable even under UV stress and desiccation. These results highlight two very important aspects: (i) our lack of understanding of and the predictability between the presence of generic E. coli and Salmonella and (ii) the different risk patterns that could be associated with water source, time of application, and physiological stage of the crop. It is unknown whether these two conditions impact the survival of enteric pathogens at different physiological stages of multiple crops and whether the VBNC state in crop protection sprays, pesticide applications, or under field conditions is impacting enteric pathogen survival and persistence.

Soils, Manure, and Biological Soil Amendments

When describing the presence, survival, persistence, and proliferation of human pathogens in soils, manure, biological soil amendments, and a combination of these, it is important to distinguish the potential contribution of each source to the overall level of contamination within each matrix. In general, soil is not considered to be a significant source of enteric pathogens for fresh produce (64), and contamination is generally associated with the presence of manure, treated or untreated biological soil amendments of animal and potentially plant origin, contaminated water, wastewater, land application of biosolids, and other anthropogenic activities (65).

Within soil, physicochemical properties including soil texture, pH, organic matter content, cation exchange capacity, porosity, and organic and inorganic nutrient sources impact the microbial ecology of all soilborne bacteria, fungi, and enteric pathogens (66, 67). In general, it has been described that soil types and lab or field experimental scales impact soil microbial population dynamics, with contradicting results between the influence of sandy soil, loam, and clay loam on pathogen survival (13, 68–72, 178). Soil moisture, pH, and organic matter content have been suggested to be more predictive of microbial community composition and enteric pathogen survival in soil (71, 73–76).

Soil microbial community interactions with enteric pathogens could be a meaningful factor in understanding the survival, persistence, and growth of these pathogens. There have been contradicting studies that either support or refute the idea that greater soil microbial diversity will negatively impact enteric pathogen survival and persistence, with little to no information on the impact on growth of enteric pathogens in soil (13, 69, 70). Among these studies, the most complete evaluation of the impact of microbial communities on the survival and persistence of E. coli O157:H7 was performed by van Elsas et al. (13). In this study, an inverse relationship between soil microbial diversity and survival of E. coli O157:H7 was established in fabricated microcosms. This was indicative of resistance to invasion due to greater soil species diversity and decreased ability to compete for resources by E. coli O157:H7. These results are not supported by reports from Ibekwe et al. (69) and Gagliardi and Karns (70), who used other generic and pathogenic strains of E. coli in their studies. Results from these experiments suggest that other microbial species influence these interactions, and further studies are needed to elucidate if nutrient acquisition, predation, or inhibition impacts enteric pathogen survival and persistence. Interactions of this nature are especially important with other human pathogens such as L. monocytogenes, which tends to behave like a true soil microorganism. Recently, Delgado-Baquerizo et al. (7) published the first comprehensive genome list of the 500 most common true soil microorganisms across multiple soil types and physicochemical characteristics. This information would help in the development of microcosm, mesocosm, and potentially, field studies in which the interactions of these 500 soil phylotypes with enteric pathogens could be compared to determine the impact on nutrient acquisition, predation, or inhibition and the survival, persistence, and growth of enteric pathogens in soils.

The contributions of soil and air to the spread and dissemination of foodborne infectious diseases may be considered low to minimal (77). Key to this statement is our ability to determine which human pathogens or foodborne pathogens are true soil microorganisms and which are periodic, transient, or incidental soil colonizers. True soil microorganisms capable of infecting humans through food consumption are microbes capable of completing their entire life cycle in soil. Periodic, transient, or incidental soil colonizers are capable of surviving in soil for prolonged periods of time without completing their entire life cycle in this matrix (6, 77). L. monocytogenes is one of those microorganisms that is considered to have an unusually broad ecological niche and host range (78) and that has been reported to exist as a “soil resident” of decomposing organic matter (68) and has been found on multiple animal hosts (67, 79). Human outbreak investigations have demonstrated the ability of L. monocytogenes to survive and persist under quite a remarkable range of conditions (57) due to the presence of multiple gene products that facilitate the utilization of multiple carbon and other nutrient resources (79, 80). This environmental saprophyte is capable of remaining in soil and other environments for prolonged periods of time and even of growing in cold and other harsh environments (67). Its infectious or intercellular life cycle is typically completed inside a host cell; however, based on its known ability to grow in an array of environmental conditions, including soil, can we consider L. monocytogenes to be a true soil microorganism? It is crucial to understand how L. monocytogenes and other enteric pathogens interact with native soil microbial communities. This information must be gathered through comprehensive studies with multiple soils in microcosm, mesocosm, and open field environments to develop a broad and detailed characterization of how soil impacts the short- and long-term survival of enteric pathogens.

Environmental and management factors that typically introduce enteric pathogens to soil include the use of raw manure, wastewater, human biosolids, compost, and wild and domestic animal intrusion. Raw manure has been used for thousands of years as a soil amendment to increase or maintain the soil nitrogen and carbon balance and to increase the aggregate stability of agricultural soils. The presence and persistence of enteric pathogens in raw and composted manure has been studied extensively, and we currently have a good understanding of what types of pathogens can be found on different animal excreta (81–83). A significant number of studies also looked at the survival of these pathogens if manure is applied to different soil types and at different rates and correlated it with their survival and persistence, which seems to fluctuate between 30 to over 350 days depending on the soil type and source of manure (84–86). Based on these results, it is important to question three important aspects of the presence of enteric pathogens in soil, manure, compost, and a combination of these. First, once manure is applied to soil, results suggest that enteric pathogen survival and persistence are enhanced (86), probably because of the addition of nutrients that are important for pathogen survival. Second, it is not clear if the addition of manure to soil overrides the resistance to invasion of soil microbial communities by enteric pathogens irrespective of greater soil species diversity and decreased competitiveness for resources as suggested by van Elsas et al. 2012 (13). Finally, properly composted manure has been shown to significantly reduce the presence of enteric pathogens (85, 87–89). It is not clear if this process also impacts the transfer of antimicrobial resistance (AMR) from thermotolerant coliforms and other heat-resistant microorganisms such enterococci to human pathogens present in compost or soil. These pressing questions will significantly impact the viability and continuity of the national organic program 90-120 rule for the use of raw or aged manure incorporated into the soil, since numerous results indicate that this approach may not provide the necessary level of protection to reduce contamination of the edible portion of the crop (84–86). Currently, several national studies are looking at the impact of this rule on public health and how proper composting practices impact the continuation of this standard or the development of new guidelines for the application of raw, aged, and composted manure at specific crop intervals that will significantly reduce produce contamination. Lastly, we are currently facing significant challenges at all scales of production and within our health system, because of the wide spread of AMR microorganisms. With this in mind, we lack understanding on how different approved composting methods impact inactivation and/or transfer of AMR between organisms and to food if compost is used as a source of nutrients for plant growth. Further studies looking at this conundrum are needed to determine how antibiotic use in animals and composting processes are impacting fruit and vegetable contamination with microorganisms presenting broad resistance to different antimicrobials.

Wildlife and Domestic Animals

Contamination of fresh produce with enteric pathogens has been associated with fecal contamination coming from wild and domestic animals (4, 27, 90). Prevention of fecal contamination from wildlife has been suggested to be the most effective approach to reducing contamination of fresh produce; however, controlling wildlife intrusion on farms in some instances is not feasible, especially when dealing with birds, small mammals, and insects. The prevalence of enteric pathogens in the environment and wildlife populations plays a significant role in the levels of contamination in fresh produce, especially when buffer zones and wildlife corridors are removed from agricultural farmland (11). Gorski et al. (10) investigated the prevalence of enteric pathogens in produce environments and wildlife around the Salinas Valley, and found different levels of Salmonella contamination in produce, water, and wild animals during the cropping cycle, suggesting that these pathogens were persistent in the environment. Subsequently, losses in wildlife habitat coupled with the natural prevalence of enteric pathogens in wildlife, will significantly impact the level of contamination in fresh produce. The prevalence of many enteric pathogens on farmland or wildlife across the United States is unknown, and this information is needed for proper risk assessment of produce contamination and to determine if remediation practices are available and viable to reduce pathogen contamination.

Proximity to small or large confined-animal operations has also been suggested to be a potential source of enteric pathogen contamination due to aerosol contamination and water runoff when produce fields are in close proximity to these operations. Berry et al. (91) and Jahne et al. (92), discovered aerosol distribution of enteric microorganisms from large animal operations and manure/dust to fresh produce up to distances of over 150 meters, suggesting that the extent of produce contamination is greater in locations where large-scale animal production is adjacent to produce fields. No conclusive evidence of foodborne pathogen transfer or human illness is available that links the proximity of animal operations, size of animal operations, aerosols dispersal, and pathogen contamination with a known foodborne outbreak; however, results from these experiments and recent work by Kumar et al. (93) in tomatoes, suggest that aerosols are a plausible source of contamination. One key component missing from these evaluations is the scale of the animal operation and whether it impacts pathogen transmission. Many organic and diversified farming systems utilize work animals or source manure from small herds within the farm to fertilize their fields with raw, aged, or composted manure. It is unknown what the level of enteric pathogen contamination is in these small-scale animal systems and whether aerosols and proximity to these animal activities represents a significant risk of produce contamination.

Packinghouse, Equipment, and Tools

Outbreak investigations of multiple crops, including melons, watermelons, apples, stone fruit, and other fruit and vegetables (4, 57, 58, 90, 94, 95) have been linked to contaminated packinghouses or equipment used to handle and transport produce. Contamination of equipment or areas within the packinghouse has also been suggested to originate from field contamination, poor sanitation practices, traffic flow, worker health and hygiene, or the overall design of the facility (4, 5, 95). Irrespective of the source of contamination, numerous questions remain unanswered about the fitness of enteric pathogens in those environments, under stress conditions, low populations, and the VBNC stage and how sanitizers impact survival, persistence, and in some cases, resistance to these chemical substances either through biofilm formation or pathogen adaptation (96, 97). The 2014 caramel apple outbreak associated with listeriosis left a significant imprint in our current knowledge on the adaptability of L. monocytogenes to a wide range of environments. It became clear that this pathogen could persist in dry packinghouse environments for long periods of time under conditions previously not known to support the survival and persistence of this pathogen (57). During the investigation of this outbreak, over 110 samples were collected from multiple locations, and seven samples were positive for L. monocytogenes, six of them coming from food contact surfaces including polishing brushes, drying brushes, a conveyor, and inside a wooden bin. In general, all these locations are difficult to clean and sanitize because of poor design of facilities and equipment, highlighting gaps in our understanding of microbiological risks associated with tree fruit production, especially activities related to maintaining, cleaning, and sanitizing of food contact surfaces and how biofilms may persist for prolonged periods of time under suboptimal conditions (dry and low nutrient availability). Similar observations have been identified for stone fruit (58) and melon packinghouses (95) and have triggered significant changes to industry practices.

Traffic flow within packinghouses also plays a significant role in spreading filth and enteric pathogens in different areas across the facility (5). Typical vehicles for distribution of these contaminants include pallet jacks, pallets, and worker and visitor traffic. One peculiar practice in many facilities is to move visitors from dirty to clean areas when describing the different activities that take place in the facility. This simple and yet influential practice could transfer contamination from receiving areas to finished product (clean) areas if no further steps in the process are available to reduce new contamination events. In a recent study performed in multiple food facilities in Europe by Muhterem-Uyar et al. (98), researchers found three major sources/areas of contamination. These were reception of raw materials, processing areas, and widespread contamination of different microorganisms across the processing facilities. Although these studies were performed on processing facilities instead of fresh produce packinghouses, results also point out traffic flow as one of the major factors impacting the distribution of contamination.

Similar studies on fresh produce packinghouses or fresh-cut conditions are limited. Erickson (8) followed outbreak investigations of cabbage, carrots, celery, onions, and deli salads and found that at these facilities and irrespective of geographical region, transfer of contamination could also be associated with traffic flow. Consequently, further studies looking at the transfer of contaminants across packinghouse environments due to traffic flow are necessary, especially where linear flows within the facility are not possible and where movement of pallet jacks, pallets, and other equipment is not controlled. It is also unknown how these conditions impact the survival and persistence of enteric pathogens in low concentrations (<100 cells) and whether different food-grade lubricants facilitate pathogen persistence on equipment and tools.

Sanitation of packinghouse equipment and tools continues to be a major source of misunderstanding and confusion within different scales of fruit and vegetable production. At the core of the problem lie four major factors: (i) facility and equipment design, (ii) lack of understanding of the proper use of chemical sanitizers and of the development of sound sanitation practices, (iii) new-buyer guidelines that in some instances push for environmental monitoring plans that are not required under the FSMA Produce Safety Rule, and (iv) a lack of technical background in the development, implementation, and assessment of environmental monitoring plans. A good example that highlights some of these deficiencies was observed in a multistate outbreak associated with Salmonella enterica serotype Newport linked to mangoes (94). During the investigation of this outbreak, it was determined that a lack of sanitation practices, coupled with the use of hot water treatments without sanitizers and water coming from canals from a river 26 km away from the packinghouse were the major factors linked to the source of contamination. Since this outbreak, the U.S. mango industry has worked on updating and developing new cleaning and sanitation practices and updating equipment and other tools that could reduce or eliminate similar problems (99).

The proper use of sanitizers, including adjustments of pH, temperature, and turbidity when appropriate; monitoring tools; and intervals that can guarantee the effectiveness of the system are scarce in the production of many fruits and vegetables, except for fresh tomato flume systems (19, 63, 100). Numerous outbreaks associated with enteric pathogens (mainly Salmonella) in tomatoes have generated the need to establish mitigation practices that could reduce pathogen contamination (4, 101). One of these mitigation practices targets packinghouse sanitation and the correct use of a “fruit disinfection treatments”, since postharvest washing was once considered a potential decontamination step in the process instead of an activity to reduce cross contamination between fruit loads and is now clearly identified as a high-risk practice when performed incorrectly. In general, tomato flume systems maintain adequate chlorine levels in the dump tank to prevent pathogen survival, transfer, and internalization of the fruit. However, continuous fluctuations in chlorine concentrations are observed due to rapid accumulation of organic matter that reacts with free chlorine, causing a significant decline in sanitizer concentrations and leaving washing systems vulnerable to pathogen survival and persistence. For these reasons, maintaining chlorine concentrations and its effectiveness relies on maintaining the water and tomato pulp temperatures within 5.5°C and a contact time of no longer than 2 minutes (100). These parameters depend on numerous factors, including the load of debris on the fruit, the quantity of fruit dumped in the tanks, the speed with which the fruit is moved inside the tanks, and the ability of the system and operators to react to constant changes (100). Similar factors can also determine the effectiveness of apple, cherry, pear, melon, watermelon, mango, papaya, and other dump tank systems and highlights the inherit difficulty that many growers are currently facing with their washing systems. To date, several laboratory methods and parameters have been reported that could be used for the sanitation of packinghouse and crop disinfection treatment systems, but further research is needed to fully understand how chlorine or other sanitizers effectively reduce human pathogen contamination, especially when penetration of protective sites on fruit and vegetable surfaces is limited by water tension properties of the sanitizer solution.

PRODUCTION PRACTICES IMPACTING FRESH PRODUCE SAFETY

The wide diversity and size of farm management systems looking for ways to simplify certain farm practices to increase yields and lower costs through fertilizer and pesticide applications, combined with the use and design of equipment, environmental conditions, and federal, state, buyer, and consumer demands within regional and local markets impacts the safety of fresh produce. Farm management systems vary with the farming practices (organic, conventional sustainable, diversified) employed during the growing season, and the microbial risks associated with each system and its crops are linked to grower, packer, shipper, and handler practices in different geographical locations and personal business philosophies. Nonetheless, fruit and vegetable farming systems and their impact on fresh produce safety can be separated based on farming practices.

Conventional versus Organic Farming Systems

At the core of these two systems lies the fundamental difference in the use of synthetic fertilizers and pesticides to promote plant growth and control the incidence of pests. This simple and yet fundamental difference between farming activities mainly impacts farming equipment, seed sources, resistance to specific pests, crop diversity, postharvest disinfection, and some farm-related activities including the timing, method, and frequency of the application of different materials and substances used for pest control and crop fertilization. Conventional farming systems tend to be monocultures since the use of fertilizers and chemical pesticides allows for proper nutrient and pest management. However, organic systems cannot rely on these inputs and follow diversified farming principles that allow them to better control pests and potentially manage nitrogen if nitrogen-fixing cover crops are used within the distribution of crops. Aside from these variances, any other inputs, including water sources, harvest crews, transportation, handling, and storage conditions, of conventional and organic produce tend to be almost identical. To date, very few studies have compared the two production systems in terms of their microbial characteristics and the safety of fresh produce (102, 103).

Hoogenboom et al. (104) evaluated the level of contaminants and microorganisms in Dutch organic and conventional food products and found that they had similar scores with regard to food safety; however, the levels of antibiotic-resistant microorganisms were lower in organic systems. Human pathogen outbreaks have been linked to more conventional than organic produce, but these differences could be explained by the sheer volume of conventional produce sold each year compared to organic produce. Further, results from Williams and Hammit (105), Magkos et al. (106), and Hoogenboom et al. (104) support the idea that there seem to be no significant differences in the incidence of enteric pathogens between organic and conventional systems. Looking at the main differences between farming systems, two practices can significantly influence the presence, survival, persistence, and growth of enteric pathogens and impact the risk of contamination of fresh produce in organic systems: (i) the use of raw, aged, and composted manure and the time and quantity of these amendments applied either to the soil or through the irrigation systems (as in the case of compost teas, fish emulsion, and other substances) and (ii) the use of acetic acid (organic origin), alcohol (organic origin), chlorine, hydrogen peroxide, and peroxyacetic acid in organic packinghouse environments, which have been registered for organic packinghouse surface sanitation, while only ozone, chlorine, and peroxyacetic acid have been registered for fruit and vegetable disinfection (107). How these sanitizers impact the survival and persistence of enteric pathogens and the microbial communities within these systems and whether these parameters are different in conventional packinghouse systems is unknown. Essential oils from thyme, lemongrass, eucalyptus, rosemary, oregano, and sweet basil have also been used to sanitize food contact surfaces and for fresh produce disinfection. However, their use is limited due to transfer of flavor to produce or food contact surfaces, since effective antimicrobial doses exceed acceptable organoleptic levels in produce (108).

Sustainable Farming Systems

Sustainable farming, at its core, strives to combine and utilize the best practices from organic and conventional systems that will allow the rational use of farming inputs while maintaining yields and reducing chemical inputs into the environment. This type of farming system could also utilize raw, aged, and composted manure and farm animals to reduce carbon footprints within the system. To date, there is no clear understanding of whether these farming systems pose equal, lower, or higher risk factors for enteric pathogenic contamination and how fresh produce safety is impacted. Similar to organic systems, the main risk associated with some sustainable farming practices is linked to the proximity of animals and the use of raw or aged manure to fertilize crops. Concerns similar to those for organic practices exist specific to the presence, survival, and growth of enteric pathogens and the impact of small animal herds on the transfer of these pathogens through aerosols or runoff. Leff and Fierer’s (102) bacterial community analysis of organic and conventional produce yielded significant differences in the relative abundance of the Enterobacteriaceae taxa in organic and conventional produce, with the latter presenting a higher presence of this taxa than organic produce. These differences could be attributed to produce source since most foods were purchased in retail establishments, where further handling and processing occurs. Despite this potentially important difference for food safety, overall, based on their results, consumers are exposed to a substantially different bacterial community when consuming organic produce. Similar observations were made by Maffei et al. (103) when looking at published data describing microbial composition and contamination of organic and conventional produce from different countries. Whether these differences are important for risk assessment of foodborne illnesses and transfer of AMR has yet to be defined, and further efforts are needed to understand these potential similarities.

Greenhouses

Fruit and vegetable greenhouse production exceeded $3 billion in sales in 2013, and it is projected to continue growing and surpass $4 billion in the next 3 to 5 years. The biggest incremental growth in production (2-fold) has been associated with leafy greens and culinary herbs, while fruit production has almost tripled since 2009 (14). Leafy green, culinary herb, and fruit greenhouse production systems utilize different water sources and hydroponic systems; however, in most of them, nutrient solution recirculation is a common practice that minimizes water use, improves nitrogen use, and in some instances reduces pesticide applications (109). Recirculation of the nutrient solution could potentially become a food safety hazard, depending on the microbial quality of the water used for irrigation, human contamination, contaminated substrates used for seed germination, and plant growth (110). Few greenhouse operations currently monitor the microbial quality of water associated with the contamination of human pathogens, and most efforts center on controlling plant pathogens. To date, only one clear outbreak associated with hydroponically grown cucumbers has been investigated by the CDC, and the investigation indicated poor sanitation and agricultural practices as potential sources of contamination. The report also looked into whether cross-contamination could have occurred during handling, distribution, and transport of the fruit; however, this association was not established (2).

Numerous studies have also looked at the potential of fruit or leaf internalization of enteric pathogens through the root system in different crop physiological stages. Although some evidence has been reported, most of the studies (except for tomato [111]) support that little to no known internalization and translocation of enteric pathogens from the root system to leaves or other organs has been established (73, 112–115). Consequently, in hydroponic systems, special attention to reduce or eliminate enteric pathogen contamination should focus on preventing these pathogens from entering greenhouse operations through sick workers, poor sanitation practices, contaminated substrates, and the use of surface or rain water sources that could carry enteric pathogens into the system.

SUSTAINABLE PRACTICES TO MITIGATE FOOD SAFETY RISKS ON PRODUCE FARMS

To date, there is a clear absence and need for the development of sustainable mitigation practices that can reduce or eliminate enteric pathogen contamination from farmland within a relatively short timeframe for the long-term sustainability and stability of the produce industry. Mitigation techniques that follow sound and sustainable farming practices are potential alternatives to chemical-based remediation methods that could eventually reduce soil health and crop production (116, 117). Short-term cover cropping, compost application, and solarization have been proposed as potential alternatives to reduce pathogen and weed contamination from soils while maintaining soil health and crop productivity (118–121). The key to the use and implementation of these techniques is cost, effectiveness, and the ability to implement them across a wide range of cropping systems.

Efforts toward the development of practices that reduce contamination from multiple sources have focused on reducing runoff from confined-animal feedlot operations (CAFOs) and on preventing intrusion from wildlife and domestic animals into production areas (5, 122, 123). Runoff water from feedlots and pastures can be a major source of microbial contamination of surface water and produce farms (5). Potential environmental risk factors associated with the 2007 spinach outbreak were traced back to several factors, including previous land history, the presence of wild pigs, and irrigation wells near surface waterways exposed to feces from cattle and wildlife. The outbreak strain E. coli O157:H7 was identified in river water, cattle feces, and wild pig feces on the ranch and within a mile of the implicated field (90).

Comanagement of food safety practices that will reduce enteric pathogen contamination of produce with environmental policies, the economy, and consumer expectations remains a significant challenge across the farm to fork continuum. One of the biggest challenges in this process relates to the limited number of studies looking at this topic at different levels in the food system and the difficulty of extrapolating specific or local results to multiple regions and cropping systems. Here, we present a number of sustainable practices that could be implemented within comanagement of farm food safety practices and that have been suggested to reduce the microbial food safety hazard at the farm level.

Buffer Zones

There are a number of different definitions of buffer zones, but in general terms and based on National Organic Program regulations, a buffer zone is “an area located between a certified production operation or portion of a production operation and an adjacent land area that is not maintained under organic management” (124). It also refers to a strip of land adjacent to production areas or bodies of water. In general, its purpose is to protect farmland from prohibited substances in the case of organic farming systems or to protect waterbodies from pollutants, excess sediments, and microbial contaminants that may negatively impact water quality. Buffer zones also serve as a source of food, nesting cover, and shelter for many wildlife species and can serve as wildlife corridors that could minimize wildlife intrusion into production areas (125, 126). Buffer zones may be natural land where existing vegetation is intact, or they can be developed by altering the slope of the land and adding vegetation and trees. Vegetable production areas that are in close proximity to CAFOs are at higher risk of microbial contamination directly or indirectly from animals, runoff water, or other vectors associated with these types of operations. Dillaha et al. (127), suggested that prevention of produce field contamination from runoff from CAFOs could be achieved by constructing diversion ditches and establishing vegetation barriers such as riparian buffers, filter strips, and grassed waterways.

Several studies have looked at the impact of buffer zones on reducing or augmenting the transmission of human pathogens from CAFOs, animal grazing lands, and large bodies of water (128–131). These studies suggest that buffer zones tend to have a protective effect against the transmission of human pathogens to produce fields. Recently, Karp et al. (11) looked in the Salinas Valley at the impact of removing buffer zones from adjacent produce farmland on the incidence of human pathogens in the finished product. The study found that pathogen prevalence increased the most on farms where noncrop vegetation was removed. These results suggest the importance of potentially keeping or expanding these zones, especially in growing areas in which intense wildlife pressure or proximity to large-scale animal operations poses a significant risk of contamination. It is worth noting that these buffer zones, when located between CAFOs and farmland, tend to accumulate dust and other particles coming from the animal units, and it is not known whether under those conditions buffer zones could become sinks and sources of contamination that could spread enteric pathogens to adjacent farmland.

Vegetative Filter Strips (Buffer Strips)

Vegetative filter strips (VFSs) are another type of buffer zone and refers to lands with a variety of grasses and forage species located downslope from the potential source of contaminants and upslope from the area being protected (132–134). The main purpose of installing VFSs is to protect surface water quality by reducing the level of physical, chemical, and microbial pollutants in agricultural runoff (135). Several studies reported that VFSs can effectively reduce the level of coliforms, E. coli (136, 137), nitrogen, and phosphorus (138) in waterways that comes from contaminated runoff from the source of pollutants (181).

VFSs are installed by planting a band of vegetation of a single species or a mixture of grasses, legumes, and/or other forbs, also known as herbaceous flowering plants not including graminoids (127, 135). Once a band of vegetation is installed, it works naturally. The vegetation interrupts overland flow and reduces surface runoff velocity. The reduced flow allows the surface runoff to sit on the soil for a longer period of time, increasing the rate of deposition of sediments, nutrients, and human pathogens. It also increases the rate of infiltration into the soil and adsorption of sediments and some nutrients on leaves and stems. The plant roots and soil microorganisms degrade nutrients and chemical pollutants (139). All these mechanisms collectively help reduce the level of surface water pollution through contaminated surface run-off.

Water flowing through the VFS decreases in velocity, and this consequently decreases the sediment-carrying capacity of the runoff as particles settle (140). The nutrients and microorganisms which are attached to sediment particles are retained in the VFS (127). Conflicting results have been reported on the effectiveness of VFSs in minimizing microbial runoff. Some studies indicated that VFSs are effective in trapping nutrients/solid particles and reducing microbial runoff (91% for fecal coliform bacteria and 74% for fecal streptococci) from pasture or feedlot areas (136, 141–143), while a few others have found limited beneficial effects in reducing fecal contamination in swine wastewater moving through vegetative areas regardless of vegetation type or season (137, 140, 144–146).

Variable results have been reported on the effectiveness of VFSs in removing the pollutants in feedlot runoff manure-applied pasture and cropland runoff (140). Factors impacting the efficiency of these systems include the amount and type of incoming pollutant, slope, runoff volume, type of flow (concentrated or diffusive), and type of vegetation (147, 148). Table 2 describes some of the major pollutants and important parameters within VFSs that improve filtration. There should be an adequate level of plant density for the efficient retention of pollutants. Infiltration is one of the main mechanisms for removing soluble pollutants. However, an increase in the duration of flow may result in decreases in the rate of removal of pollutants and could explain why in some evaluations, the effectiveness of reducing pathogen contamination from waterways was not observed. Overall, these VFSs have shown significant potential for pathogen removal. Questions remain as to their effectiveness over time, especially with sediment accumulation due to weather events or continued water movement from different farm areas. It has also been suggested that over time, these areas could become sinks and sources of contamination for different pollutants, but their impact on human pathogen accumulation has not been well described.

TABLE 2.

Examples of VFSs evaluated for eliminating the pollutants

Vegetation	Major pollutants	Source	VFS parameters	References
Kikuyu grass (Pennisetum clandestinum Chiov) and Napier grass (Pennisetum purpureum Schumach)	E. coli	Cattle manure (cowpat)	VFS slope: 15%VFS length: 44 mVFS width: 4 m	175
Fescue (Festuca)	Fecal coliforms, Streptococci, E. coli, nitrogen, phosphorus	Feedlot	VFS slope: 2%VFS length: 30 mVFS width: 15 mSoil type: Newtonia silt loam soil	176
Mixed grass buffer strip	Phosphorus, ammonium, nitrogen, potassium	Feedlot	VFS slope: 2%VFS length: 12 m	140
Blue fescue (Festuca ovina L. ‘Glauca’) and white clover (Trifolium repens L.) mixture	Fecal coliforms	Manure	VFS slope: 20%VFS length: 6 mVFS width: 2 mSoil type: sandy loam or clay loam	177

Soil Solarization

Solarization is a convenient nonchemical, nonfumigant sustainable farming technique that utilizes daily solar heating cycles to manage weeds, nematodes, diseases, and insects in soil. Depending on the region, this process is also known as polyethylene mulching, soil trapping, solar pasteurization, and solar soil heating. Soil solarization reduces the survival and persistence of mesophilic microorganisms and plant pathogens within the first 4 to 5 cm of the soil profile by raising the soil temperature to over 40°C (120, 180). Critical to achieving these temperatures is the application of water, keeping soil moisture contents above field capacity but below saturation, and applying this technique during the summer months when longer photoperiods (>11 h), elevated temperatures (>30°C), and higher UV indexes increase the efficiency of this process (149–154).

Survival of enteric pathogens in soil for an extended period of time has been observed under multiple experimental conditions (22, 86, 155). However, survival may be affected when the organisms are exposed to adverse environmental conditions such as limited nutrients and exposure to UV light. Every organism has an optimum temperature range for growth. At temperatures above 50°C the growth of most enteric pathogens is affected, and die-off is expedited. Soil solarization practices are capable of reaching these temperatures and could potentially reduce the population of enteric pathogens and other transient soil colonizers. To achieve significant inactivation of pathogens, weeds, and other diseases, several farm practices should be implemented, including soil ripping and tilling, incorporation of organic matter including compost, which will increase aeration, conductivity and heat transfer via mechanisms of exothermic microbial activity, and thermal conductivity within the top 20 to 40 cm of the soil profile (152, 154). An effective solarization process, in addition to reducing plant pathogens, weeds, and other disease, is conducive to increased nutrient availability, plant growth, and beneficial changes in the population dynamics of soil bacteria and fungi (121, 154, 156, 157), which makes this an excellent sustainable practice to maintain soil health that potentially could reduce the presence of enteric pathogens. To that end, soil solarization has been reported to be effective in reducing generic E. coli in soil (119) and in feedlot pen surface material (158) after a 6-week treatment, suggesting that this practice could potentially be used to control other human pathogens, including Salmonella and L. monocytogenes.

Short-Term Cover Cropping

Enteric pathogen (Salmonella spp., E. coli O157:H7, Shigella spp., and Campylobacter jejuni) survival and persistence in soil has been reported to depend on temperature (159, 160), soil moisture content (73, 159), soil physicochemical properties (161, 162), soil C:N ratio (75), microbial community composition (13, 163), and UV light exposure (164). Prolonged survival of enteric pathogens in soils threatens the sustainability of farming because there are currently no remediation practices to remove them from contaminated soil, and the current industry best practice is to abandon these fields and move to other growing regions (117). Consequently, developing and implementing soil remediation practices that reduce or eliminate human pathogen contamination is crucial for the long-term sustainability of the produce industry.

Soil physicochemical properties, nutrient composition, and the microbial communities in the rhizosphere impact enteric pathogen survival and are impacted by short-term cover cropping. The growth of cover crops is an important agricultural management practice that is widely applied in different cropping systems to improve soil health and plant productivity (165). Numerous types of plants can be used as cover crops; however, legumes and grasses are extensively used to improve nitrogen availability and scavenge for excess soil nutrients, in particular nitrogen, and to reduce soil erosion (166). There is increasing interest in Fabaceae (sunn hemp), Brassica (rape, mustard, and forage radish), and Polygonaceae (buckwheat) cover crops because of their allelopathic effects on weeds and plant diseases (166).

In addition to solarization, short-term cover cropping has been proposed as an alternative mitigation practice to reduce human pathogen contamination from soil because of its ability to be implemented across a wide range of cropping systems and the known allelopathic effects due to the action of secondary plant metabolites on soil disease management (152, 153). The proximity of fresh produce to contaminated soil and enteric pathogen concentrations are two risk factors contributing to produce contamination (167). Generally, produce that is in close proximity to the soil (spinach, lettuce, root crops, melon) has a higher likelihood of contamination compared to produce that develops further above the soil surface (tomatoes, peppers, berries) (16, 17). Contaminated irrigation water and manure are thought to be two important sources of enteric pathogen contamination in fresh produce (128, 129).

Sunn hemp, buckwheat, and mustard greens are three short-term and low-residue cover crops that produce distinct secondary plant metabolites, which have been shown to exhibit bactericidal properties against human pathogens (110, 168, 169). The availability of these plant secondary metabolites to inactivate human pathogens in soil is dependent on the ability to macerate and incorporate plant tissue through disking and tilling after growing the cover crops for 30 to 60 days. Sunn hemp (Crotalaria juncea) produces pyrrolizidine alkaloids, which are nitrogen-containing compounds that can be toxic to animals after consumption (170). However, no reports to date have documented their inhibitory effects on enteric pathogen growth in soils or other matrixes.

Buckwheat (Fagopyrum esculentum) mainly produces two phenolic compounds: rutin and chlorogenic acid. Rutin in its pure form has not shown inhibitory properties against enteric pathogens (169, 171, 172, 179). Marginal inhibitory effects have been observed when it is mixed with quercetin, suggesting a synergistic inhibitory effect between the substances at a concentration of 100 μg/ml (171). Chlorogenic acid in its pure form has shown inhibitory effects against S. aureus, Salmonella, and E. coli at concentrations of 40 to 80 ppm (169).

Mustard green cover crops (Brassica juncea) produce glucosinolates, which are sulfur-containing water-soluble compounds that when hydrolyzed by the myrosinace enzyme produce isothiocyanate compounds known to have inhibitory properties against E. coli, S. enterica serotype Typhimurium, and L. monocytogenes in concentrations ranging from 50 to 200 ppm (117, 168, 173, 174). Allyl isothiocyanate, in a liquid form, has also been shown to inhibit growth of E. coli, S. Typhimurium, and L. monocytogenes in concentrations ranging from 50 to 200 ppm (110). Altogether, combining the use of cover cropping, composting, and solarization could potentially be an alternative and sustainable practice capable of remediating soil contaminated with enteric pathogens. The applicability of this method in commercial organic and conventional fresh produce farming systems is not known.