- Email: [email protected]
Incidence of naturally internalized bacteria in lettuce leaves
International Journal of Food Microbiology 162 (2013) 260–265
Contents lists available at SciVerse ScienceDirect
International Journal of Food Microbiology journal homepage: www.elsevier.com/locate/ijfoodmicro
Incidence of naturally internalized bacteria in lettuce leaves Zhe Hou a, Ryan C. Fink a, Christie Radtke a, Michael J. Sadowsky b, c, Francisco Diez-Gonzalez a,⁎ a b c
Department of Food Science and Nutrition, University of Minnesota, St. Paul, MN 55108, USA BioTechnology Institute, University of Minnesota, St. Paul, MN 55108, USA Department of Soil, Water, and Climate, University of Minnesota, St. Paul, MN 55108, USA
a r t i c l e
i n f o
Article history: Received 29 August 2012 Received in revised form 23 January 2013 Accepted 28 January 2013 Available online 13 February 2013 Keywords: Natural internalization Bacteria Lettuce leaves Endophytes
a b s t r a c t Lettuce is the fresh leafy vegetable most frequently involved in foodborne disease outbreaks. Human bacterial pathogens may be experimentally internalized into lettuce plants, but the occurrence of natural microflora inside lettuce leaves has not been elucidated. To characterize the endophytic microorganism residing in commercial lettuce leaves, two separate studies were conducted. First, a total of 30 and 25 heads of romaine and red leaf lettuce, respectively, served as the source of individual leaves which were surface sterilized, stomached, enriched in BHI broth for 24 h and plated onto BHI agar for non-selective isolation of internalized microorganism. In a separate survey, 80 heads of each of the two types of lettuce were similarly processed, except that GN broth and MacConkey agar (MCA) were used for isolation of Gram negative bacteria. Thirty-eight out of 100 leaves were positive for internalized microorganisms, and Bacillus, Pseudomonas and Pantoea were the genera most frequently found in both types of lettuce. Members of the genus Erwinia were isolated from romaine lettuce only. In the second study, 21 and 60% of romaine and red leaf lettuce heads, respectively, had internalized bacteria capable of growing on MCA. Among the Gram negative strains, Pseudomonas and Pantoea genera were most frequently isolated. Enterobacter isolates were obtained from three red leaf samples. In summary, spore-forming bacteria and traditional epiphytic bacterial genera were frequently detected in surface-sterilized commercial lettuce leaves. Despite the common occurrence of internalized bacteria, only Enterobacter was related to Escherichia coli O157:H7 and Salmonella. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Fresh produce such as lettuce and spinach has been recognized as important vehicles for the transmission of human pathogens (Berger et al., 2010). Salmonella enterica and Escherichia coli strain O157:H7 are the two pathogenic bacteria that have caused most of the recent produce outbreaks. Contamination of leafy green vegetables can occur at many stages of production as these pathogens can cycle through the environment and food chain via manure, insects, water, soil, and food (Semenov et al., 2010). Lettuce appears to be particularly susceptible to E. coli O157 and Salmonella contamination. Franz et al. (2007) reported that 18% of all lettuce-associated outbreaks were caused by Salmonella and 10% of all Salmonella outbreaks with fresh produce were related to lettuce. The lettuce-associated outbreaks linked to E. coli O157:H7 were more than double that of Salmonella. Recently, it has been suggested that bacterial pathogens may be found inside of lettuce leaves. It has been hypothesized that these human pathogens might get internalized into lettuce plants through wounds or via roots and stomata (Erickson et al., 2010c; Klerks et al., ⁎ Corresponding author at: Department of Food Science and Nutrition, University of Minnesota, 1334 Eckles Ave, St. Paul, MN 55108, United States. Tel.: +1 612 624 9756; fax: +1 612 625 5272. E-mail address: [email protected] (F. Diez-Gonzalez). 0168-1605/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.ijfoodmicro.2013.01.027
2007a, 2007b). The internalized cells would survive typical surface produce washing and disinfection. The reported uptake of E. coli O157: H7 by the roots of growing lettuce plants (Solomon and Matthews, 2005; Semenov et al., 2009) and of Salmonella by hydroponic tomato and lettuce plants (Guo et al., 2002; Klerks et al., 2007a, 2007b) led to the hypothesis that foodborne pathogens can exist as endophytes within growing plants (Gu et al., 2011). Research undertaken by Solomon et al. (2002) and Klerks et al. (2007a, 2007b) further supported the idea that E. coli O157:H7 can enter lettuce plants through the root system and migrate to the edible portion of the plant. Other internalization routes, in addition to the roots such as the entrance through the stomata or via cuts and wounded leaf edges, have also been investigated (Takeuchi and Frank, 2000; Takeuchi et al., 2000). Erickson et al. (2010a) reported that the internalization of E. coli O157:H7 into lettuce leaves could be enhanced by insect and physical damage. Another study suggested that vacuum cooling might significantly increase the infiltration of E. coli O157 into lettuce leaves (Li et al., 2008). Although results from some studies suggested that human pathogens could internalize into plant tissues, quantitative information on the prevalence and level of infection have not been investigated (Franz et al., 2007). Moreover, some investigators did not find internalization of Salmonella or E. coli O157 into lettuce plants, despite the fact that the experiments were conducted under laboratory conditions using environmentally unrealistic concentrations of bacteria (Erickson et al., 2010b; Franz et al., 2007; Zhang et al., 2009b). Despite these
Z. Hou et al. / International Journal of Food Microbiology 162 (2013) 260–265
efforts, the natural incidence of internalization is yet to be confirmed. Some studies have already begun to investigate the natural incidence of bacterial internalization in rice, maize, tomato and wheat (Rosenblueth and Martinez-Romero, 2006; Cevallos-Cevallos et al., 2012a, 2012b) However, to our knowledge no other study has been conducted to investigate natural internalization in lettuce. In the study reported here, using an efficient sterilization method (ethanol and sodium hypochlorite plus UV light) for the surface of lettuce leaves, the incidence of internalized aerobic and Gram-negative bacteria in commercial leaves of romaine and red leaf lettuce was determined (Hallmann et al., 1997). The internalized bacteria in lettuce leaves were identified by molecular and biochemical methods. 2. Materials and methods 2.1. Sampling method and leaf sterilization Romaine and red leaf lettuce heads were purchased randomly on a weekly basis from at least 4 different supermarkets in the metropolitan area of St. Paul, MN, U.S.A. Samples were stored at 4 °C and tested within 48 h after purchase. Four to five outer leaves were removed by hand and the inner leaves from heads of fresh romaine and red leaf lettuce were surface sterilized by immersion in an 80% ethanol solution for 10 s, followed by immersion in 2% sodium hypochlorite for 30 min. Leaves were kept in continuous contact with the sterilant by use of a stir bar and leaves were finally rinsed three times, in sterile de-ionized water. Leaves were further sterilized by using a UV light (190–290 nm) for 30 min, on each side. Sterilization efficacy was evaluated by placing randomly selected leaves directly onto the surface of brain heart infusion (BHI; BD, Franklin Lakes, NJ) agar plates or MacConkey agar plates (MCA; Neogen, Inc., Lansing, MI); leaves were overlaid with liquefied, cooled, BHI agar or MacConkey agar and incubated at 37 °C for 3 days. 2.2. Isolation on different media The remainder of the sterilized leaves which weigh from 2 to 5 g was individually homogenized with a stomacher (Seward, Ontario) and enriched in 99 ml BHI broth and incubated for 24 h at 37 °C. Enriched cultures were streaked with a loop onto BHI agar plates and incubated for 24 h at 37 °C for single colony isolation. Secondly, we also hypothesized that commercial red leaf and romaine lettuce leaves contained internalized Gram-negative bacteria such as E. coli. To test this hypothesis, 11 g aliquots of leaves were blended with 99 ml phosphate buffered saline (PBS), pH 7.0, and directly plated on MacConkey agar (MCA; Neogen, Inc., Lansing, MI) and incubated for 24 h at 37 °C for total bacterial counts on MCA. The remainders of leaves from each head were surface sterilized as described above and blended in a stomacher in 99 ml Gram negative (GN) broth (BD, Franklin Lakes, NJ). Cultures were incubated at 37 °C for 24 h and streaked onto MCA plates, which were incubated for 24 h at 37 °C for single colony isolation.
261
kits (Valencia, CA). The concentration of DNA was measured spectrophotometrically at 260 nm. A 40 to 60 pg aliquot of each purified DNA was sequenced at the University of Minnesota BioMedical Genomics Center. For this initial sequencing the size of the 5′ fragments ranged from 256 to 340 bp and for the 3′ PCR products was from 170 to 311 bp. The sequence data were analyzed by using Chromas Lite ver. 2.01 (Eden Prairie, MN). Based on this analysis the genus of isolates was determined. DNA sequence alignments of 5′ and 3′ regions of 16S rRNA genes were performed by using the Clustal V program (Higgins and Sharp, 1989). The SEQBOOT program (Felsenstein, 1985) was used to generate 100 bootstrap replicates for each alignment. Phylogenies based on unrooted parsimony were generated using the PHYLIP package (Felsenstein, 1985). Distance matrices of aligned sequences were generated using the DNADIST program and the distances were converted into phylogenetic trees using the Neighbor program (Saitou and Nei, 1987). The CONSENSE program was used to construct majority rule consensus trees. 2.4. Identification by Biolog plates and API strips Biolog analyses were used to identify repeated isolates from the BHI plates and reduce the number of isolates subjected to complete sequencing of the 16S rRNA genes. Biolog GN2 and GP plates (Hayward CA) were inoculated and incubated following the manufacturer's directions (http://www.biolog.com/pdf/GN2b_Brochure.pdf, http://www.biolog. com/pdf/GP2b_Brochure.pdf) and color reactions in wells were visually recorded and measured at 600 nm by using a microplate spectrophotometer. Results were transferred to binary data using a positive cut off O.D.600 nm =0.374 (if X>0.374, 1, 0). The Biolog binary data profiles were analyzed using a simple matching method and Bionumerics ver.3.5 (Applied Maths, Sint-Maratens-Latem, Belgium). For clusters of strains having a similarity value greater than 90%, only one isolate was chosen to determine the complete sequence of the 16S rRNA amplicon (size range 1511 to 1553 bp). In contrast, all isolates having a similarity value less than 90% in a cluster were chosen for full sequence analysis. The isolates from the same cluster of the phylogenetic tree were also grouped based on data obtained using Biolog GN2 and GP2 plates and Bionumerics software. Strains obtained from MCA were taxonomically identified by using API-20E® strips (bioMérieux, St Louis, MO) following the manufacturer's instructions. No 16S rRNA sequencing was used for these isolates. 2.5. Statistical analysis Statistical analyses were carried-out to determine the significance of the differences in prevalence of total count of Gram-negative bacteria between red leaf and romaine lettuce by using Odds Ratio (P ≤ 0.05). The same statistical analysis was also applied to determine whether there were any differences between the prevalence of internalized bacteria in red leaf and romaine lettuce.
2.3. 16S rRNA sequencing and phylogenetic trees
3. Results
Isolates obtained from BHI plates were initially classified by 16S rRNA sequencing as described by Lane (1991). A three-digit code (A–B–C) for each isolate was used, with A being the batch number; B, the head number; and C, the leaf number. For example, 8–1–3 indicates this isolates was taken from batch 8, head 1, and was number 3 leaf. Furthermore, the character “O” before the three letter code meant the isolate was from romaine lettuce; the “E” character meant that the isolate was from red leaf lettuce. The universal primers used in this study were: 8F: 5′-AGAGTTTGATCCTGGCTCAG-3′; 1492R: 5′GGTTACCTTGTTACGACTT-3′. The PCR products were electrophoresed in 0.8% agarose gels and were purified using QIAquick PCR purification
The method chosen for sterilization, treatment with 80% ethanol for 10 s, followed by immersion in 2% NaOCl for 30 min and 30 min drying under UV light for each side of lettuce leaves proved to be effective in killing surface-associated bacteria. For every sampling batch, two leaves of each lettuce type were randomly chosen following surface sterilization and fully covered in sterile, cooled, BHI or MacConkey agar. After three days of incubation at 37 °C, none of the tested leaves showed bacterial growth (data not shown). In the first part of the study, a total of 30 and 25 romaine and red leaf lettuce heads, respectively served as the source of individual leaves (from 3 to 4 leaves per head). From 100 surface sterilized romaine
262
Z. Hou et al. / International Journal of Food Microbiology 162 (2013) 260–265
lettuce leaves, 38 leaves (38%) obtained from 14 (47%) heads were found to contain internalized bacteria (Table 1). Initial sequence analysis of 16S rRNA flanking fragments from 38 of these isolates indicated that 82% (31 of 38) of the microorganisms present inside romaine lettuce were Bacillus. Gram negative bacteria belonging to the genera Pseudomonas (8%), Erwinia (8%) and Pantoea (2%) were also recovered from inside of romaine lettuce leaves (Table 2). The same trend was also found for red leaf lettuce. Of 100 leaves examined from 25 heads, 38 leaves (38%) from 16 (64%) heads were found positive for internalized bacteria (Table 1). Sequence analysis of 16S rRNA flanking fragments from these bacteria indicated that 22 of these isolates were Bacillus (58%), while 14 and 2 strains were Pseudomonas (37%) and Pantoea (5%) genera, respectively (Table 2). The 16S rRNA gene sequences of all the isolates from romaine and red leaf lettuce were analyzed and clustered phylogenetically (Figs. 1 and 2). Not surprisingly, four clusters of bacteria were found in trees constructed from isolates obtained from romaine and red lettuce leaves. Generally, isolates from the same batch and head fell into the same cluster (Fig. 1). In order to reduce the number of strains to be examined by complete 16S rRNA analysis of the PCR amplicon, isolates from the same clusters in the phylogenetic tree were examined for their biochemical profiles using Biolog GN or GP plates. After analyzing the Biolog data, a 90% similarity value of those strains was chosen as our cut off number to reduce the number of strains for fully sequencing of the 16S rRNA PCR product. For instance, if strains fell into same cluster and those strains had a similarity equal or larger than 90%, one of them was randomly picked. In consequence, the 16S rRNA amplicon of six and 10 Gram-negative and -positive isolates was fully sequenced (Fig. 3A and B). Eighty heads of each romaine and red leaf lettuce were analyzed for the presence of total and internalized Gram negative bacteria. The total bacterial count on non-sterilized leaves was greater in red lettuce than in romaine lettuce (Fig. 4). In particular, the incidence of total bacteria counts greater than 1000 CFU/g of red leaf lettuce heads (38%) was significantly more frequent than with romaine lettuce (6.3%) samples (P b 0.05) (Table 3). The same general trend was also observed for internalized bacteria although there was no significant difference between red leaf and romaine lettuce (Table 3). The number of surface sterilized red leaf lettuce heads positive for bacteria was three-fold greater than romaine lettuce heads. As many as 48 out of 80 surface sterilized read leaf lettuce heads (60%) were found to contain internalized bacteria. In contrast, 21% of romaine lettuce heads (17 out of 80) were found positive for internalized bacteria after MCA enrichment. Among the strains which could be identified for both types of lettuce, Pantoea was the most frequently isolated Gram negative bacterium (59% and 56% of isolates from romaine and red leaf lettuce respectively). Pseudomonas was also isolated for both types of lettuce and a small percentage of Enterobacter isolates (6%) was only obtained from three red leaf samples (Table 4). 4. Discussion The natural occurrence and persistence of foodborne pathogens on fresh produce are often ascribed to a shift of these microorganisms to an endophytic lifestyle (Solomon et al., 2002). It has been speculated that once internalized, the bacteria normally susceptible to the harsh
Table 1 Number of lettuce heads, and leaves that had naturally internalized microorganisms using non-selective enrichment and plating media (brain heart infusion broth and agar). Lettuce cultivars
Number of heads
Number of leaves
Number of positive heads
Number of positive leaves
Romaine Red leaf
30 25
100 100
14 (47%) 16 (64%)
38 (38%) 38 (38%)
Table 2 Genera distribution of bacterial isolates recovered from inside surface-sterilized lettuce leaves identified by 16S rRNA sequencing. Isolation was performed using a combination of non-selective enrichment and plating (BHI broth and agar). Genus
Bacillus Pseudomonas Erwinia Pantoea
Number of isolates from: Romaine
Red leaf
31 (82%) 3 (8%) 3 (8%) 1 (2%)
22 (58%) 14 (37%) 0 2 (5%)
environmental conditions on the leaf surface can find shelter and survive once they gain access to the internal plant tissues. Several studies have focused on the internalization of E. coli O157:H7 and Salmonella in lettuce plants (Erickson et al., 2010a; Takeuchi and Frank, 2000; Zhang et al., 2009b; Klerks et al., 2007a, 2007b). Indeed, a recent study reported that human pathogens such as E. coli O157:H7 and Salmonella can get internalized into lettuce plants and these internalized cells survive post-washing steps (Erickson et al., 2010c). Unfortunately, all of these studies have used highly concentrated bacteria inocula that may occur rarely in real life conditions (Solomon et al., 2002; Takeuchi and Frank, 2000). Salmonella and Escherichia belong to the Enterobacteriaceae family (Franz et al., 2007). Several members of this family are either plant endophytes and/or plant pathogens such as species belonging to the Erwinia and Pantoea genera (Rosenblueth and Martinez-Romero, 2006). Because of the relative taxonomic closeness between plant and human Enterobacteriaceae, it could be hypothesized that the latter may also be able to invade plant tissues. Internalization may be a complex natural process that allows bacteria to enter the plant through different routes. Hence, it is plausible that E. coli and Salmonella can enter the plant through similar routes. As mentioned before, although there are several studies on internalization in artificially inoculated lettuce, to our knowledge there are no studies focusing on natural internalization in lettuce. This study is the first to investigate and describe the incidence of naturally internalized bacteria in lettuce. One of the major challenges to determine the rate of internalization of bacteria is preventing contamination of the sample with microflora present on the surface of the leaves. Therefore, it is necessary to ensure their complete elimination. There are several sterilization methods reported in the literature. For instance, Zhang et al. (2009a) and Franz et al. (2007) evaluated and summarized a list of sanitizers proven to kill human pathogens efficiently. Their studies determined that immersion in 80% ethanol for 10 s, followed by immersion in 0.1% HgCl2 for 10 min and 1% of AgNO3 solution for 10 s, was the most effective in inactivating E. coli O157:H7, Salmonella and Listeria monocytogenes on lettuce surfaces. In our evaluation, this method failed to completely remove the flora present on the leaf surface (data not shown). Most likely this occurred because the natural microbiota present on lettuce leaves is more resilient to sterilization than the human pathogens tested by Zhang et al. (2009a). The use of sodium hypochlorite for disinfecting plant surfaces is also a common and well characterized sterilization method (Hallmann et al., 1997). Our modification of this treatment by including an additional rinse in 80% ethanol for 10 s, followed by immersion in 2% NaOCl for 30 min and 30 min drying under UV light for each side of the leaf was effective to fully sterilize the surface of both romaine and red leaf lettuce. Our analysis with non-selective medium (BHI) showed that, after enrichment of bacteria from surface sterilized leaves, Gram negative genera present in the lettuce plants were Pantoea, Erwinia, and Pseudomonas. This finding corroborates other studies that found that these bacteria were present as plant endophytes in rice and soybean as well (Rosenblueth and Martinez-Romero, 2006). Although the definition of endophyte describes a microorganism that can be isolated from surface-sterilized plant tissue and, generally, does not
Z. Hou et al. / International Journal of Food Microbiology 162 (2013) 260–265
263
Fig. 1. Phylogenetic tree of naturally internalized microorganisms from romaine lettuce. Each three-digit code represents an individual isolate. Three digit code for each isolate (A–B–C) means A: batch number, B: head number and C: leaf number.
cause any visible harm to the plant, there are reports that under some environmental conditions these bacteria can turn into plant pathogens (Rosenblueth and Martinez-Romero, 2006). We also found a relatively high prevalence of Bacillus species internalized in lettuce indicating these isolates can survive, and colonize well inside lettuce leaves. This is not surprising as the presence of this genus has also been previously reported in other plants, including sugar beets and potato. Bacillus thuringiensis, which has the ability to protect the crop from different insects by producing δ-endotoxin (Hallmann et al., 1997), was also isolated from romaine lettuce in our study. Since E. coli O157:H7 and Salmonella enterica belong to the Enterobacteriaceae family (Franz et al., 2007), we also used a specific selection technique to increase the chances of obtain isolates from
lettuce leaves belonging to this family. Moreover, to determine if there is any significant difference and relationship between the microflora present on the surface and inside the leaves, we measured the total bacteria population by direct plate counting. Interestingly, we found that red leaf lettuce had a significantly larger population of total bacteria than romaine lettuce, and prevalence of the internalized Enterobacteriaceae for red leaf lettuce was much higher than romaine lettuce. This might indicate that the prevalence of naturally internalized bacteria varies between plant cultivars. Indeed, Pillay and Nowak (1997) also reported that the endophytic colonization on tomato plants differed from one cultivar to another. Based on previous studies showing internalization of leaf-applied bacteria by using addition of highly concentrated inocula, it is possible that the greater
Fig. 2. Phylogenetic tree of naturally internalized microorganisms from red leaf lettuce. Each three-digit code represents an individual isolate. Three digit code for each isolate (A–B–C) means A: batch number, B: head number and C: leaf number.
264
Z. Hou et al. / International Journal of Food Microbiology 162 (2013) 260–265
Fig. 3. Comparison of Biolog data of Gram-negative (A) and Gram-positive (B) naturally internalized microorganisms for both romaine and red leaf lettuce. Matching bacterial species and strain number were obtained from the Ribosomal Database Project (http://rdp.cme.msu.edu/seqmatch/seqmatch_intro.jsp).
likelihood of recovering internalized bacteria from inside of the leaves is dependent on a larger surface bacteria population (Pillay and Nowak, 1997). Similar to what we found in our initial survey, Pantoea and Pseudomonas were recovered from inside of lettuce leaves. Interestingly, Enterobacter sp. strains were only isolated from red leaf lettuce. This bacterium has also been isolated as an endophyte in cucumber and corn (Hallmann et al., 1997). Additionally, Cooley et al. (2003) found that Enterobacter asburiae was able to eliminate S. enterica and E. coli O157:H7 from Arabidopsis thaliana seeds which suggested that this serovar could potentially be used as a biocontrol agent for human pathogens.
One of the few studies that have characterized the microbiota associated with lettuce was recently published by Rastogi et al. (2012). Using 16SRNA pyrosequencing some of the major bacterial genera found in romaine lettuce were Pseudomonas, Bacillus, Massilia, Arthrobacter and Pantoea. Another report characterizing the population diversity and structure also found Pseudomonas, Erwinia, Enterobacter, and Pantoea as the major genera in lettuce plants (Hunter et al., 2010). Despite the fact that none of those publications targeted internalized bacteria, it should be noted that several of the bacterial genera were consistent with those found in our investigation. Table 3 Prevalence of total count and internalized bacteria between red leaf and romaine lettuce. Category
Total bacteria count (>1000 CFU/g) Red leaf Romaine Incidence of internalization Red leaf Romaine
Fig. 4. Total counts of Gram negative bacteria from non-treated romaine and red leaf lettuce leaves.
% positive
37.5 (30/80)c 6.3 (5/80)
60 (48/80) 21 (17/80)
Odds ratioa
11
3.5
Confidence interval 95%b Lower
Upper
9.9
12.1
0.5
1.9
a Odds ratio defines the risk (odds) of finding a positive sample relative to the category with an odds ratio value of 1. b Significant variable when confidence interval for the odds ratio does not contain 1.00. c The number was calculated based on the total number of lettuce heads containing Gram negative bacteria.
Z. Hou et al. / International Journal of Food Microbiology 162 (2013) 260–265 Table 4 Genera distribution of naturally internalized bacteria recovered from lettuce leaves by enrichment in GN broth and plating onto MacConkey agar. Genera were determined by API-20E® testing. Bacterial genera
Pantoea Pseudomonas Enterobacter Unknown
Number of isolates from Romaine
Red leaf
10 (59%)a 2 (12%) 0 5 (29%)
27 (56%) 6 (13%) 3 (6%) 12 (25%)
a The number was calculated based on the total number of lettuce heads containing naturally internalized Gram negative bacteria (17 for romaine and 48 for red leaf lettuce).
In this study, we did not find any E. coli or Salmonella either inside or outside of both types of lettuce, although Salmonella has been reported to be found in alfalfa sprouts (Ponka et al., 1995) and is frequently associated with lettuce outbreaks (Franz et al., 2007). This is not surprising, as Franz et al. (2008) calculated that the probability of finding E. coli O157:H7 inside of lettuce heads was 1:10,000. Clearly further studies are needed to address the role that these bacteria play within plants. While some studies have suggested that endophytes may regulate the plant defense system and this likely influences their colonization status inside of plants (Iniguez et al., 2005; Klerks et al., 2007a, 2007b), this currently remains speculative. 5. Conclusion In this study, we sampled lettuce heads and found that sporeforming endophytic bacteria were easily detected in a large number of surface-sterilized commercial lettuce leaves. These data indicate that bacterial endophytes are commonly present in lettuce plants. Moreover, very few of the bacterial genera were related to Gram-negative pathogenic bacteria transmitted by leafy greens, such as E. coli O157:H7 and Salmonella. Acknowledgment Special thanks to John Ferguson for analyzing the Biolog binary data. This project was funded by a grant from the University of Minnesota's Healthy Foods, Healthy Lives Institute. References Berger, C.N., Sodha, S.V., Shaw, R.K., Griffin, P.M., Pink, D., Hand, P., Frankel, G., 2010. Fresh fruit and vegetables as vehicles for the transmission of human pathogens. Environmental Microbiology 12, 2385–2397. Cevallos-Cevallos, J.M., Danyluk, M.D., Gu, G.Y., Vallad, G.E., van Bruggen, A.H.C., 2012a. Dispersal of Salmonella Typhimurium by rain splash onto tomato plants. Journal of Food Protection 75, 472–479. Cevallos-Cevallos, J.M., Gu, G.Y., Danyluk, M.D., Dufault, N.S., van Bruggen, A.H.C., 2012b. Salmonella can reach tomato fruits on plants exposed to aerosols formed by rain. International Journal of Food Microbiology 158, 140–146. Cooley, M.B., Miller, W.G., Mandrell, R.E., 2003. Colonization of Arabidopsis thaliana with Salmonella enterica and enterohemorrhagic Escherichia coli O157:H7 and competition by Enterobacter asburiae. Applied and Environmental Microbiology 69, 4915–4926. Erickson, M.C., Liao, J., Payton, A.S., Riley, D.G., Webb, C.C., Davey, L.E., Kimbrel, S., Ma, L., Zhang, G.D., Flitcroft, I., Doyle, M.P., Beuchat, L.R., 2010a. Preharvest internalization of Escherichia coli O157:H7 into lettuce leaves, as affected by insect and physical damage. Journal of Food Protection 73, 1809–1816. Erickson, M.C., Webb, C.C., Diaz-Perez, J.C., Phatak, S.C., Silvoy, J.J., Davey, L., Payton, A.S., Liao, J., Ma, L., Doyle, M.P., 2010b. Infrequent internalization of Escherichia coli O157: H7 into field-grown leafy greens. Journal of Food Protection 73, 500–506. Erickson, M.C., Webb, C.C., Diaz-Perez, J.C., Phatak, S.C., Silvoy, J.J., Davey, L., Payton, A.S., Liao, J., Ma, L., Doyle, M.P., 2010c. Surface and internalized Escherichia coli O157:H7 on field-grown spinach and lettuce treated with spray-contaminated irrigation water. Journal of Food Protection 73, 1023–1029.
265
Felsenstein, J., 1985. Confidence-limits on phylogenies-an approach using the bootstrap. Evolution 39, 783–791. Franz, E., Visser, A.A., Van Diepeningen, A.D., Klerks, M.M., Termorshuizen, A.J., van Bruggen, A.H.C., 2007. Quantification of contamination of lettuce by GFP-expressing Escherichia coli O157:H7 and Salmonella enterica serovar Typhimurium. Food Microbiology 24, 106–112. Franz, E., Semenov, A.V., van Bruggen, A.H.C., 2008. Modelling the contamination of lettuce with Escherichia coli O157:H7 from manure-amended soil and the effect of intervention strategies. Journal of Applied Microbiology 105, 1569–1584. Gu, G.Y., Hu, J.H., Cevallos-Cevallos, J.M., Richardson, S.M., Bartz, J.A., van Bruggen, A.H.C., 2011. Internal colonization of Salmonella enterica serovar Typhimurium in tomato plants. PloS One 6, e27340. Guo, X.A., van Iersel, M.W., Chen, J.R., Brackett, R.E., Beuchat, L.R., 2002. Evidence of association of Salmonellae with tomato plants grown hydroponically in inoculated nutrient solution. Applied and Environmental Microbiology 68, 3639–3643. Hallmann, J., Quadt-Hallmann, A., Mahaffee, W.F., Kloepper, J.W., 1997. Bacterial endophytes in agricultural crops. Canadian Journal of Microbiology 43, 895–914. Higgins, D.G., Sharp, P.M., 1989. Fast and sensitive multiple sequence alignments on a microcomputer. Computer Applications in the Biosciences 5, 151–153. Hunter, P.J., Hand, P., Pink, D., Whipps, J.M., Bending, G.D., 2010. Both leaf properties and microbe-microbe interactions influence within-species variation in bacterial population diversity and structure in the lettuce (Lactuca species) phyllosphere. Applied and Environmental Microbiology 76, 8117–8125. Iniguez, A.L., Dong, Y.M., Carter, H.D., Ahmer, B.M.M., Stone, J.M., Triplett, E.W., 2005. Regulation of enteric endophytic bacterial colonization by plant defenses. Molecular Plant-Microbe Interactions 18, 169–178. Klerks, M.M., Franz, E., van Gent-Pelzer, M., Zijlstra, C., van Bruggen, A.H.C., 2007a. Differential interaction of Salmonella enterica serovars with lettuce cultivars and plant-microbe factors influencing the colonization efficiency. The ISME Journal 1, 620–631. Klerks, M.M., van Gent-Pelzer, M., Franz, E., Zijlstra, C., van Bruggen, A.H.C., 2007b. Physiological and molecular responses of Lactuca sativa to colonization by Salmonella enterica serovar Dublin. Applied and Environmental Microbiology 73, 4905–4914. Lane, D.J., 1991. 16S/23S rRNA sequencing. In: Stackebrandt, E., Goodfellow, M. (Eds.), Nucleic Acid Techniques in Bacterial Systematics. John Wiley & Son Ltd., U.K., pp. 115–147. Li, H.P., Tajkarimi, M., Osburn, B.I., 2008. Impact of vacuum cooling on Escherichia coli O157:H7 infiltration into lettuce tissue. Applied and Environmental Microbiology 74, 3138–3142. Pillay, V.K., Nowak, J., 1997. Inoculum density, temperature, and genotype effects on in vitro growth promotion and epiphytic and endophytic colonization of tomato (Lycopersicon esculentum L) seedlings inoculated with a pseudomonad bacterium. Canadian Journal of Microbiology 43, 354–361. Ponka, A., Andersson, Y., Siitonen, A., Dejong, B., Jahkola, M., Haikala, O., Kuhmonen, A., Pakkala, P., 1995. Salmonella in alfalfa sprouts. Lancet 345, 462–463. Rastogi, G., Sbodio, A., Tech, J.J., Suslow, T.V., Coaker, G.L., Leveau, J.H., 2012. Leaf microbiota in an agroecosystem: spatiotemporal variation in bacterial community composition on field-grown lettuce. The ISME Journal 6, 1812–1822. http:// dx.doi.org/10.1038/ismej.2012.32. Rosenblueth, M., Martinez-Romero, E., 2006. Bacterial endophytes and their interactions with hosts. Molecular Plant-Microbe Interactions 19, 827–837. Saitou, N., Nei, M., 1987. The neighbor-joining method-a new method for reconstructing phylogenetic trees. Molecular Biology and Evolution 4, 406–425. Semenov, A.V., van Overbeek, L., van Bruggen, A.H.C., 2009. Percolation and survival of Escherichia coli O157:H7 and Salmonella enterica serovar Typhimurium in soil amended with contaminated dairy manure or slurry. Applied and Environmental Microbiology 75, 3206–3215. Semenov, A.M., Kuprianov, A.A., van Bruggen, A.H., 2010. Transfer of enteric pathogens to successive habitats as part of microbial cycles. Microbial Ecology 60, 239–249. Solomon, E.B., Matthews, K.R., 2005. Use of fluorescent microspheres as a tool to investigate bacterial interactions with growing plants. Journal of Food Protection 68, 870–873. Solomon, E.B., Yaron, S., Matthews, K.R., 2002. Transmission of Escherichia coli O157:H7 from contaminated manure and irrigation water to lettuce plant tissue and its subsequent internalization. Applied and Environmental Microbiology 68, 397–400. Takeuchi, K., Frank, J.F., 2000. Penetration of Escherichia coli O157:H7 into lettuce tissues as affected by inoculum size and temperature and the effect of chlorine treatment on cell viability. Journal of Food Protection 63, 434–440. Takeuchi, K., Matute, C.M., Hassan, A.N., Frank, J.F., 2000. Comparison of the attachment of Escherichia coli O157:H7, Listeria monocytogenes, Salmonella typhimurium, and Pseudomonas fluorescens to lettuce leaves. Journal of Food Protection 63, 1433–1437. Zhang, G.D., Ma, L., Beuchat, L.R., Erickson, M.C., Phelan, V.H., Doyle, M.P., 2009a. Evaluation of treatments for elimination of foodborne pathogens on the surface of leaves and roots of lettuce (Lactuca sativa L.). Journal of Food Protection 72, 228–234. Zhang, G.D., Ma, L., Beuchat, L.R., Erickson, M.C., Phelan, V.H., Doyle, M.P., 2009b. Lack of internalization of Escherichia coli O157:H7 in lettuce (Lactuca sativa L.) after leaf surface and soil inoculation. Journal of Food Protection 72, 2028–2037.
Contents lists available at SciVerse ScienceDirect
International Journal of Food Microbiology journal homepage: www.elsevier.com/locate/ijfoodmicro
Incidence of naturally internalized bacteria in lettuce leaves Zhe Hou a, Ryan C. Fink a, Christie Radtke a, Michael J. Sadowsky b, c, Francisco Diez-Gonzalez a,⁎ a b c
Department of Food Science and Nutrition, University of Minnesota, St. Paul, MN 55108, USA BioTechnology Institute, University of Minnesota, St. Paul, MN 55108, USA Department of Soil, Water, and Climate, University of Minnesota, St. Paul, MN 55108, USA
a r t i c l e
i n f o
Article history: Received 29 August 2012 Received in revised form 23 January 2013 Accepted 28 January 2013 Available online 13 February 2013 Keywords: Natural internalization Bacteria Lettuce leaves Endophytes
a b s t r a c t Lettuce is the fresh leafy vegetable most frequently involved in foodborne disease outbreaks. Human bacterial pathogens may be experimentally internalized into lettuce plants, but the occurrence of natural microflora inside lettuce leaves has not been elucidated. To characterize the endophytic microorganism residing in commercial lettuce leaves, two separate studies were conducted. First, a total of 30 and 25 heads of romaine and red leaf lettuce, respectively, served as the source of individual leaves which were surface sterilized, stomached, enriched in BHI broth for 24 h and plated onto BHI agar for non-selective isolation of internalized microorganism. In a separate survey, 80 heads of each of the two types of lettuce were similarly processed, except that GN broth and MacConkey agar (MCA) were used for isolation of Gram negative bacteria. Thirty-eight out of 100 leaves were positive for internalized microorganisms, and Bacillus, Pseudomonas and Pantoea were the genera most frequently found in both types of lettuce. Members of the genus Erwinia were isolated from romaine lettuce only. In the second study, 21 and 60% of romaine and red leaf lettuce heads, respectively, had internalized bacteria capable of growing on MCA. Among the Gram negative strains, Pseudomonas and Pantoea genera were most frequently isolated. Enterobacter isolates were obtained from three red leaf samples. In summary, spore-forming bacteria and traditional epiphytic bacterial genera were frequently detected in surface-sterilized commercial lettuce leaves. Despite the common occurrence of internalized bacteria, only Enterobacter was related to Escherichia coli O157:H7 and Salmonella. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Fresh produce such as lettuce and spinach has been recognized as important vehicles for the transmission of human pathogens (Berger et al., 2010). Salmonella enterica and Escherichia coli strain O157:H7 are the two pathogenic bacteria that have caused most of the recent produce outbreaks. Contamination of leafy green vegetables can occur at many stages of production as these pathogens can cycle through the environment and food chain via manure, insects, water, soil, and food (Semenov et al., 2010). Lettuce appears to be particularly susceptible to E. coli O157 and Salmonella contamination. Franz et al. (2007) reported that 18% of all lettuce-associated outbreaks were caused by Salmonella and 10% of all Salmonella outbreaks with fresh produce were related to lettuce. The lettuce-associated outbreaks linked to E. coli O157:H7 were more than double that of Salmonella. Recently, it has been suggested that bacterial pathogens may be found inside of lettuce leaves. It has been hypothesized that these human pathogens might get internalized into lettuce plants through wounds or via roots and stomata (Erickson et al., 2010c; Klerks et al., ⁎ Corresponding author at: Department of Food Science and Nutrition, University of Minnesota, 1334 Eckles Ave, St. Paul, MN 55108, United States. Tel.: +1 612 624 9756; fax: +1 612 625 5272. E-mail address: [email protected] (F. Diez-Gonzalez). 0168-1605/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.ijfoodmicro.2013.01.027
2007a, 2007b). The internalized cells would survive typical surface produce washing and disinfection. The reported uptake of E. coli O157: H7 by the roots of growing lettuce plants (Solomon and Matthews, 2005; Semenov et al., 2009) and of Salmonella by hydroponic tomato and lettuce plants (Guo et al., 2002; Klerks et al., 2007a, 2007b) led to the hypothesis that foodborne pathogens can exist as endophytes within growing plants (Gu et al., 2011). Research undertaken by Solomon et al. (2002) and Klerks et al. (2007a, 2007b) further supported the idea that E. coli O157:H7 can enter lettuce plants through the root system and migrate to the edible portion of the plant. Other internalization routes, in addition to the roots such as the entrance through the stomata or via cuts and wounded leaf edges, have also been investigated (Takeuchi and Frank, 2000; Takeuchi et al., 2000). Erickson et al. (2010a) reported that the internalization of E. coli O157:H7 into lettuce leaves could be enhanced by insect and physical damage. Another study suggested that vacuum cooling might significantly increase the infiltration of E. coli O157 into lettuce leaves (Li et al., 2008). Although results from some studies suggested that human pathogens could internalize into plant tissues, quantitative information on the prevalence and level of infection have not been investigated (Franz et al., 2007). Moreover, some investigators did not find internalization of Salmonella or E. coli O157 into lettuce plants, despite the fact that the experiments were conducted under laboratory conditions using environmentally unrealistic concentrations of bacteria (Erickson et al., 2010b; Franz et al., 2007; Zhang et al., 2009b). Despite these
Z. Hou et al. / International Journal of Food Microbiology 162 (2013) 260–265
efforts, the natural incidence of internalization is yet to be confirmed. Some studies have already begun to investigate the natural incidence of bacterial internalization in rice, maize, tomato and wheat (Rosenblueth and Martinez-Romero, 2006; Cevallos-Cevallos et al., 2012a, 2012b) However, to our knowledge no other study has been conducted to investigate natural internalization in lettuce. In the study reported here, using an efficient sterilization method (ethanol and sodium hypochlorite plus UV light) for the surface of lettuce leaves, the incidence of internalized aerobic and Gram-negative bacteria in commercial leaves of romaine and red leaf lettuce was determined (Hallmann et al., 1997). The internalized bacteria in lettuce leaves were identified by molecular and biochemical methods. 2. Materials and methods 2.1. Sampling method and leaf sterilization Romaine and red leaf lettuce heads were purchased randomly on a weekly basis from at least 4 different supermarkets in the metropolitan area of St. Paul, MN, U.S.A. Samples were stored at 4 °C and tested within 48 h after purchase. Four to five outer leaves were removed by hand and the inner leaves from heads of fresh romaine and red leaf lettuce were surface sterilized by immersion in an 80% ethanol solution for 10 s, followed by immersion in 2% sodium hypochlorite for 30 min. Leaves were kept in continuous contact with the sterilant by use of a stir bar and leaves were finally rinsed three times, in sterile de-ionized water. Leaves were further sterilized by using a UV light (190–290 nm) for 30 min, on each side. Sterilization efficacy was evaluated by placing randomly selected leaves directly onto the surface of brain heart infusion (BHI; BD, Franklin Lakes, NJ) agar plates or MacConkey agar plates (MCA; Neogen, Inc., Lansing, MI); leaves were overlaid with liquefied, cooled, BHI agar or MacConkey agar and incubated at 37 °C for 3 days. 2.2. Isolation on different media The remainder of the sterilized leaves which weigh from 2 to 5 g was individually homogenized with a stomacher (Seward, Ontario) and enriched in 99 ml BHI broth and incubated for 24 h at 37 °C. Enriched cultures were streaked with a loop onto BHI agar plates and incubated for 24 h at 37 °C for single colony isolation. Secondly, we also hypothesized that commercial red leaf and romaine lettuce leaves contained internalized Gram-negative bacteria such as E. coli. To test this hypothesis, 11 g aliquots of leaves were blended with 99 ml phosphate buffered saline (PBS), pH 7.0, and directly plated on MacConkey agar (MCA; Neogen, Inc., Lansing, MI) and incubated for 24 h at 37 °C for total bacterial counts on MCA. The remainders of leaves from each head were surface sterilized as described above and blended in a stomacher in 99 ml Gram negative (GN) broth (BD, Franklin Lakes, NJ). Cultures were incubated at 37 °C for 24 h and streaked onto MCA plates, which were incubated for 24 h at 37 °C for single colony isolation.
261
kits (Valencia, CA). The concentration of DNA was measured spectrophotometrically at 260 nm. A 40 to 60 pg aliquot of each purified DNA was sequenced at the University of Minnesota BioMedical Genomics Center. For this initial sequencing the size of the 5′ fragments ranged from 256 to 340 bp and for the 3′ PCR products was from 170 to 311 bp. The sequence data were analyzed by using Chromas Lite ver. 2.01 (Eden Prairie, MN). Based on this analysis the genus of isolates was determined. DNA sequence alignments of 5′ and 3′ regions of 16S rRNA genes were performed by using the Clustal V program (Higgins and Sharp, 1989). The SEQBOOT program (Felsenstein, 1985) was used to generate 100 bootstrap replicates for each alignment. Phylogenies based on unrooted parsimony were generated using the PHYLIP package (Felsenstein, 1985). Distance matrices of aligned sequences were generated using the DNADIST program and the distances were converted into phylogenetic trees using the Neighbor program (Saitou and Nei, 1987). The CONSENSE program was used to construct majority rule consensus trees. 2.4. Identification by Biolog plates and API strips Biolog analyses were used to identify repeated isolates from the BHI plates and reduce the number of isolates subjected to complete sequencing of the 16S rRNA genes. Biolog GN2 and GP plates (Hayward CA) were inoculated and incubated following the manufacturer's directions (http://www.biolog.com/pdf/GN2b_Brochure.pdf, http://www.biolog. com/pdf/GP2b_Brochure.pdf) and color reactions in wells were visually recorded and measured at 600 nm by using a microplate spectrophotometer. Results were transferred to binary data using a positive cut off O.D.600 nm =0.374 (if X>0.374, 1, 0). The Biolog binary data profiles were analyzed using a simple matching method and Bionumerics ver.3.5 (Applied Maths, Sint-Maratens-Latem, Belgium). For clusters of strains having a similarity value greater than 90%, only one isolate was chosen to determine the complete sequence of the 16S rRNA amplicon (size range 1511 to 1553 bp). In contrast, all isolates having a similarity value less than 90% in a cluster were chosen for full sequence analysis. The isolates from the same cluster of the phylogenetic tree were also grouped based on data obtained using Biolog GN2 and GP2 plates and Bionumerics software. Strains obtained from MCA were taxonomically identified by using API-20E® strips (bioMérieux, St Louis, MO) following the manufacturer's instructions. No 16S rRNA sequencing was used for these isolates. 2.5. Statistical analysis Statistical analyses were carried-out to determine the significance of the differences in prevalence of total count of Gram-negative bacteria between red leaf and romaine lettuce by using Odds Ratio (P ≤ 0.05). The same statistical analysis was also applied to determine whether there were any differences between the prevalence of internalized bacteria in red leaf and romaine lettuce.
2.3. 16S rRNA sequencing and phylogenetic trees
3. Results
Isolates obtained from BHI plates were initially classified by 16S rRNA sequencing as described by Lane (1991). A three-digit code (A–B–C) for each isolate was used, with A being the batch number; B, the head number; and C, the leaf number. For example, 8–1–3 indicates this isolates was taken from batch 8, head 1, and was number 3 leaf. Furthermore, the character “O” before the three letter code meant the isolate was from romaine lettuce; the “E” character meant that the isolate was from red leaf lettuce. The universal primers used in this study were: 8F: 5′-AGAGTTTGATCCTGGCTCAG-3′; 1492R: 5′GGTTACCTTGTTACGACTT-3′. The PCR products were electrophoresed in 0.8% agarose gels and were purified using QIAquick PCR purification
The method chosen for sterilization, treatment with 80% ethanol for 10 s, followed by immersion in 2% NaOCl for 30 min and 30 min drying under UV light for each side of lettuce leaves proved to be effective in killing surface-associated bacteria. For every sampling batch, two leaves of each lettuce type were randomly chosen following surface sterilization and fully covered in sterile, cooled, BHI or MacConkey agar. After three days of incubation at 37 °C, none of the tested leaves showed bacterial growth (data not shown). In the first part of the study, a total of 30 and 25 romaine and red leaf lettuce heads, respectively served as the source of individual leaves (from 3 to 4 leaves per head). From 100 surface sterilized romaine
262
Z. Hou et al. / International Journal of Food Microbiology 162 (2013) 260–265
lettuce leaves, 38 leaves (38%) obtained from 14 (47%) heads were found to contain internalized bacteria (Table 1). Initial sequence analysis of 16S rRNA flanking fragments from 38 of these isolates indicated that 82% (31 of 38) of the microorganisms present inside romaine lettuce were Bacillus. Gram negative bacteria belonging to the genera Pseudomonas (8%), Erwinia (8%) and Pantoea (2%) were also recovered from inside of romaine lettuce leaves (Table 2). The same trend was also found for red leaf lettuce. Of 100 leaves examined from 25 heads, 38 leaves (38%) from 16 (64%) heads were found positive for internalized bacteria (Table 1). Sequence analysis of 16S rRNA flanking fragments from these bacteria indicated that 22 of these isolates were Bacillus (58%), while 14 and 2 strains were Pseudomonas (37%) and Pantoea (5%) genera, respectively (Table 2). The 16S rRNA gene sequences of all the isolates from romaine and red leaf lettuce were analyzed and clustered phylogenetically (Figs. 1 and 2). Not surprisingly, four clusters of bacteria were found in trees constructed from isolates obtained from romaine and red lettuce leaves. Generally, isolates from the same batch and head fell into the same cluster (Fig. 1). In order to reduce the number of strains to be examined by complete 16S rRNA analysis of the PCR amplicon, isolates from the same clusters in the phylogenetic tree were examined for their biochemical profiles using Biolog GN or GP plates. After analyzing the Biolog data, a 90% similarity value of those strains was chosen as our cut off number to reduce the number of strains for fully sequencing of the 16S rRNA PCR product. For instance, if strains fell into same cluster and those strains had a similarity equal or larger than 90%, one of them was randomly picked. In consequence, the 16S rRNA amplicon of six and 10 Gram-negative and -positive isolates was fully sequenced (Fig. 3A and B). Eighty heads of each romaine and red leaf lettuce were analyzed for the presence of total and internalized Gram negative bacteria. The total bacterial count on non-sterilized leaves was greater in red lettuce than in romaine lettuce (Fig. 4). In particular, the incidence of total bacteria counts greater than 1000 CFU/g of red leaf lettuce heads (38%) was significantly more frequent than with romaine lettuce (6.3%) samples (P b 0.05) (Table 3). The same general trend was also observed for internalized bacteria although there was no significant difference between red leaf and romaine lettuce (Table 3). The number of surface sterilized red leaf lettuce heads positive for bacteria was three-fold greater than romaine lettuce heads. As many as 48 out of 80 surface sterilized read leaf lettuce heads (60%) were found to contain internalized bacteria. In contrast, 21% of romaine lettuce heads (17 out of 80) were found positive for internalized bacteria after MCA enrichment. Among the strains which could be identified for both types of lettuce, Pantoea was the most frequently isolated Gram negative bacterium (59% and 56% of isolates from romaine and red leaf lettuce respectively). Pseudomonas was also isolated for both types of lettuce and a small percentage of Enterobacter isolates (6%) was only obtained from three red leaf samples (Table 4). 4. Discussion The natural occurrence and persistence of foodborne pathogens on fresh produce are often ascribed to a shift of these microorganisms to an endophytic lifestyle (Solomon et al., 2002). It has been speculated that once internalized, the bacteria normally susceptible to the harsh
Table 1 Number of lettuce heads, and leaves that had naturally internalized microorganisms using non-selective enrichment and plating media (brain heart infusion broth and agar). Lettuce cultivars
Number of heads
Number of leaves
Number of positive heads
Number of positive leaves
Romaine Red leaf
30 25
100 100
14 (47%) 16 (64%)
38 (38%) 38 (38%)
Table 2 Genera distribution of bacterial isolates recovered from inside surface-sterilized lettuce leaves identified by 16S rRNA sequencing. Isolation was performed using a combination of non-selective enrichment and plating (BHI broth and agar). Genus
Bacillus Pseudomonas Erwinia Pantoea
Number of isolates from: Romaine
Red leaf
31 (82%) 3 (8%) 3 (8%) 1 (2%)
22 (58%) 14 (37%) 0 2 (5%)
environmental conditions on the leaf surface can find shelter and survive once they gain access to the internal plant tissues. Several studies have focused on the internalization of E. coli O157:H7 and Salmonella in lettuce plants (Erickson et al., 2010a; Takeuchi and Frank, 2000; Zhang et al., 2009b; Klerks et al., 2007a, 2007b). Indeed, a recent study reported that human pathogens such as E. coli O157:H7 and Salmonella can get internalized into lettuce plants and these internalized cells survive post-washing steps (Erickson et al., 2010c). Unfortunately, all of these studies have used highly concentrated bacteria inocula that may occur rarely in real life conditions (Solomon et al., 2002; Takeuchi and Frank, 2000). Salmonella and Escherichia belong to the Enterobacteriaceae family (Franz et al., 2007). Several members of this family are either plant endophytes and/or plant pathogens such as species belonging to the Erwinia and Pantoea genera (Rosenblueth and Martinez-Romero, 2006). Because of the relative taxonomic closeness between plant and human Enterobacteriaceae, it could be hypothesized that the latter may also be able to invade plant tissues. Internalization may be a complex natural process that allows bacteria to enter the plant through different routes. Hence, it is plausible that E. coli and Salmonella can enter the plant through similar routes. As mentioned before, although there are several studies on internalization in artificially inoculated lettuce, to our knowledge there are no studies focusing on natural internalization in lettuce. This study is the first to investigate and describe the incidence of naturally internalized bacteria in lettuce. One of the major challenges to determine the rate of internalization of bacteria is preventing contamination of the sample with microflora present on the surface of the leaves. Therefore, it is necessary to ensure their complete elimination. There are several sterilization methods reported in the literature. For instance, Zhang et al. (2009a) and Franz et al. (2007) evaluated and summarized a list of sanitizers proven to kill human pathogens efficiently. Their studies determined that immersion in 80% ethanol for 10 s, followed by immersion in 0.1% HgCl2 for 10 min and 1% of AgNO3 solution for 10 s, was the most effective in inactivating E. coli O157:H7, Salmonella and Listeria monocytogenes on lettuce surfaces. In our evaluation, this method failed to completely remove the flora present on the leaf surface (data not shown). Most likely this occurred because the natural microbiota present on lettuce leaves is more resilient to sterilization than the human pathogens tested by Zhang et al. (2009a). The use of sodium hypochlorite for disinfecting plant surfaces is also a common and well characterized sterilization method (Hallmann et al., 1997). Our modification of this treatment by including an additional rinse in 80% ethanol for 10 s, followed by immersion in 2% NaOCl for 30 min and 30 min drying under UV light for each side of the leaf was effective to fully sterilize the surface of both romaine and red leaf lettuce. Our analysis with non-selective medium (BHI) showed that, after enrichment of bacteria from surface sterilized leaves, Gram negative genera present in the lettuce plants were Pantoea, Erwinia, and Pseudomonas. This finding corroborates other studies that found that these bacteria were present as plant endophytes in rice and soybean as well (Rosenblueth and Martinez-Romero, 2006). Although the definition of endophyte describes a microorganism that can be isolated from surface-sterilized plant tissue and, generally, does not
Z. Hou et al. / International Journal of Food Microbiology 162 (2013) 260–265
263
Fig. 1. Phylogenetic tree of naturally internalized microorganisms from romaine lettuce. Each three-digit code represents an individual isolate. Three digit code for each isolate (A–B–C) means A: batch number, B: head number and C: leaf number.
cause any visible harm to the plant, there are reports that under some environmental conditions these bacteria can turn into plant pathogens (Rosenblueth and Martinez-Romero, 2006). We also found a relatively high prevalence of Bacillus species internalized in lettuce indicating these isolates can survive, and colonize well inside lettuce leaves. This is not surprising as the presence of this genus has also been previously reported in other plants, including sugar beets and potato. Bacillus thuringiensis, which has the ability to protect the crop from different insects by producing δ-endotoxin (Hallmann et al., 1997), was also isolated from romaine lettuce in our study. Since E. coli O157:H7 and Salmonella enterica belong to the Enterobacteriaceae family (Franz et al., 2007), we also used a specific selection technique to increase the chances of obtain isolates from
lettuce leaves belonging to this family. Moreover, to determine if there is any significant difference and relationship between the microflora present on the surface and inside the leaves, we measured the total bacteria population by direct plate counting. Interestingly, we found that red leaf lettuce had a significantly larger population of total bacteria than romaine lettuce, and prevalence of the internalized Enterobacteriaceae for red leaf lettuce was much higher than romaine lettuce. This might indicate that the prevalence of naturally internalized bacteria varies between plant cultivars. Indeed, Pillay and Nowak (1997) also reported that the endophytic colonization on tomato plants differed from one cultivar to another. Based on previous studies showing internalization of leaf-applied bacteria by using addition of highly concentrated inocula, it is possible that the greater
Fig. 2. Phylogenetic tree of naturally internalized microorganisms from red leaf lettuce. Each three-digit code represents an individual isolate. Three digit code for each isolate (A–B–C) means A: batch number, B: head number and C: leaf number.
264
Z. Hou et al. / International Journal of Food Microbiology 162 (2013) 260–265
Fig. 3. Comparison of Biolog data of Gram-negative (A) and Gram-positive (B) naturally internalized microorganisms for both romaine and red leaf lettuce. Matching bacterial species and strain number were obtained from the Ribosomal Database Project (http://rdp.cme.msu.edu/seqmatch/seqmatch_intro.jsp).
likelihood of recovering internalized bacteria from inside of the leaves is dependent on a larger surface bacteria population (Pillay and Nowak, 1997). Similar to what we found in our initial survey, Pantoea and Pseudomonas were recovered from inside of lettuce leaves. Interestingly, Enterobacter sp. strains were only isolated from red leaf lettuce. This bacterium has also been isolated as an endophyte in cucumber and corn (Hallmann et al., 1997). Additionally, Cooley et al. (2003) found that Enterobacter asburiae was able to eliminate S. enterica and E. coli O157:H7 from Arabidopsis thaliana seeds which suggested that this serovar could potentially be used as a biocontrol agent for human pathogens.
One of the few studies that have characterized the microbiota associated with lettuce was recently published by Rastogi et al. (2012). Using 16SRNA pyrosequencing some of the major bacterial genera found in romaine lettuce were Pseudomonas, Bacillus, Massilia, Arthrobacter and Pantoea. Another report characterizing the population diversity and structure also found Pseudomonas, Erwinia, Enterobacter, and Pantoea as the major genera in lettuce plants (Hunter et al., 2010). Despite the fact that none of those publications targeted internalized bacteria, it should be noted that several of the bacterial genera were consistent with those found in our investigation. Table 3 Prevalence of total count and internalized bacteria between red leaf and romaine lettuce. Category
Total bacteria count (>1000 CFU/g) Red leaf Romaine Incidence of internalization Red leaf Romaine
Fig. 4. Total counts of Gram negative bacteria from non-treated romaine and red leaf lettuce leaves.
% positive
37.5 (30/80)c 6.3 (5/80)
60 (48/80) 21 (17/80)
Odds ratioa
11
3.5
Confidence interval 95%b Lower
Upper
9.9
12.1
0.5
1.9
a Odds ratio defines the risk (odds) of finding a positive sample relative to the category with an odds ratio value of 1. b Significant variable when confidence interval for the odds ratio does not contain 1.00. c The number was calculated based on the total number of lettuce heads containing Gram negative bacteria.
Z. Hou et al. / International Journal of Food Microbiology 162 (2013) 260–265 Table 4 Genera distribution of naturally internalized bacteria recovered from lettuce leaves by enrichment in GN broth and plating onto MacConkey agar. Genera were determined by API-20E® testing. Bacterial genera
Pantoea Pseudomonas Enterobacter Unknown
Number of isolates from Romaine
Red leaf
10 (59%)a 2 (12%) 0 5 (29%)
27 (56%) 6 (13%) 3 (6%) 12 (25%)
a The number was calculated based on the total number of lettuce heads containing naturally internalized Gram negative bacteria (17 for romaine and 48 for red leaf lettuce).
In this study, we did not find any E. coli or Salmonella either inside or outside of both types of lettuce, although Salmonella has been reported to be found in alfalfa sprouts (Ponka et al., 1995) and is frequently associated with lettuce outbreaks (Franz et al., 2007). This is not surprising, as Franz et al. (2008) calculated that the probability of finding E. coli O157:H7 inside of lettuce heads was 1:10,000. Clearly further studies are needed to address the role that these bacteria play within plants. While some studies have suggested that endophytes may regulate the plant defense system and this likely influences their colonization status inside of plants (Iniguez et al., 2005; Klerks et al., 2007a, 2007b), this currently remains speculative. 5. Conclusion In this study, we sampled lettuce heads and found that sporeforming endophytic bacteria were easily detected in a large number of surface-sterilized commercial lettuce leaves. These data indicate that bacterial endophytes are commonly present in lettuce plants. Moreover, very few of the bacterial genera were related to Gram-negative pathogenic bacteria transmitted by leafy greens, such as E. coli O157:H7 and Salmonella. Acknowledgment Special thanks to John Ferguson for analyzing the Biolog binary data. This project was funded by a grant from the University of Minnesota's Healthy Foods, Healthy Lives Institute. References Berger, C.N., Sodha, S.V., Shaw, R.K., Griffin, P.M., Pink, D., Hand, P., Frankel, G., 2010. Fresh fruit and vegetables as vehicles for the transmission of human pathogens. Environmental Microbiology 12, 2385–2397. Cevallos-Cevallos, J.M., Danyluk, M.D., Gu, G.Y., Vallad, G.E., van Bruggen, A.H.C., 2012a. Dispersal of Salmonella Typhimurium by rain splash onto tomato plants. Journal of Food Protection 75, 472–479. Cevallos-Cevallos, J.M., Gu, G.Y., Danyluk, M.D., Dufault, N.S., van Bruggen, A.H.C., 2012b. Salmonella can reach tomato fruits on plants exposed to aerosols formed by rain. International Journal of Food Microbiology 158, 140–146. Cooley, M.B., Miller, W.G., Mandrell, R.E., 2003. Colonization of Arabidopsis thaliana with Salmonella enterica and enterohemorrhagic Escherichia coli O157:H7 and competition by Enterobacter asburiae. Applied and Environmental Microbiology 69, 4915–4926. Erickson, M.C., Liao, J., Payton, A.S., Riley, D.G., Webb, C.C., Davey, L.E., Kimbrel, S., Ma, L., Zhang, G.D., Flitcroft, I., Doyle, M.P., Beuchat, L.R., 2010a. Preharvest internalization of Escherichia coli O157:H7 into lettuce leaves, as affected by insect and physical damage. Journal of Food Protection 73, 1809–1816. Erickson, M.C., Webb, C.C., Diaz-Perez, J.C., Phatak, S.C., Silvoy, J.J., Davey, L., Payton, A.S., Liao, J., Ma, L., Doyle, M.P., 2010b. Infrequent internalization of Escherichia coli O157: H7 into field-grown leafy greens. Journal of Food Protection 73, 500–506. Erickson, M.C., Webb, C.C., Diaz-Perez, J.C., Phatak, S.C., Silvoy, J.J., Davey, L., Payton, A.S., Liao, J., Ma, L., Doyle, M.P., 2010c. Surface and internalized Escherichia coli O157:H7 on field-grown spinach and lettuce treated with spray-contaminated irrigation water. Journal of Food Protection 73, 1023–1029.
265
Felsenstein, J., 1985. Confidence-limits on phylogenies-an approach using the bootstrap. Evolution 39, 783–791. Franz, E., Visser, A.A., Van Diepeningen, A.D., Klerks, M.M., Termorshuizen, A.J., van Bruggen, A.H.C., 2007. Quantification of contamination of lettuce by GFP-expressing Escherichia coli O157:H7 and Salmonella enterica serovar Typhimurium. Food Microbiology 24, 106–112. Franz, E., Semenov, A.V., van Bruggen, A.H.C., 2008. Modelling the contamination of lettuce with Escherichia coli O157:H7 from manure-amended soil and the effect of intervention strategies. Journal of Applied Microbiology 105, 1569–1584. Gu, G.Y., Hu, J.H., Cevallos-Cevallos, J.M., Richardson, S.M., Bartz, J.A., van Bruggen, A.H.C., 2011. Internal colonization of Salmonella enterica serovar Typhimurium in tomato plants. PloS One 6, e27340. Guo, X.A., van Iersel, M.W., Chen, J.R., Brackett, R.E., Beuchat, L.R., 2002. Evidence of association of Salmonellae with tomato plants grown hydroponically in inoculated nutrient solution. Applied and Environmental Microbiology 68, 3639–3643. Hallmann, J., Quadt-Hallmann, A., Mahaffee, W.F., Kloepper, J.W., 1997. Bacterial endophytes in agricultural crops. Canadian Journal of Microbiology 43, 895–914. Higgins, D.G., Sharp, P.M., 1989. Fast and sensitive multiple sequence alignments on a microcomputer. Computer Applications in the Biosciences 5, 151–153. Hunter, P.J., Hand, P., Pink, D., Whipps, J.M., Bending, G.D., 2010. Both leaf properties and microbe-microbe interactions influence within-species variation in bacterial population diversity and structure in the lettuce (Lactuca species) phyllosphere. Applied and Environmental Microbiology 76, 8117–8125. Iniguez, A.L., Dong, Y.M., Carter, H.D., Ahmer, B.M.M., Stone, J.M., Triplett, E.W., 2005. Regulation of enteric endophytic bacterial colonization by plant defenses. Molecular Plant-Microbe Interactions 18, 169–178. Klerks, M.M., Franz, E., van Gent-Pelzer, M., Zijlstra, C., van Bruggen, A.H.C., 2007a. Differential interaction of Salmonella enterica serovars with lettuce cultivars and plant-microbe factors influencing the colonization efficiency. The ISME Journal 1, 620–631. Klerks, M.M., van Gent-Pelzer, M., Franz, E., Zijlstra, C., van Bruggen, A.H.C., 2007b. Physiological and molecular responses of Lactuca sativa to colonization by Salmonella enterica serovar Dublin. Applied and Environmental Microbiology 73, 4905–4914. Lane, D.J., 1991. 16S/23S rRNA sequencing. In: Stackebrandt, E., Goodfellow, M. (Eds.), Nucleic Acid Techniques in Bacterial Systematics. John Wiley & Son Ltd., U.K., pp. 115–147. Li, H.P., Tajkarimi, M., Osburn, B.I., 2008. Impact of vacuum cooling on Escherichia coli O157:H7 infiltration into lettuce tissue. Applied and Environmental Microbiology 74, 3138–3142. Pillay, V.K., Nowak, J., 1997. Inoculum density, temperature, and genotype effects on in vitro growth promotion and epiphytic and endophytic colonization of tomato (Lycopersicon esculentum L) seedlings inoculated with a pseudomonad bacterium. Canadian Journal of Microbiology 43, 354–361. Ponka, A., Andersson, Y., Siitonen, A., Dejong, B., Jahkola, M., Haikala, O., Kuhmonen, A., Pakkala, P., 1995. Salmonella in alfalfa sprouts. Lancet 345, 462–463. Rastogi, G., Sbodio, A., Tech, J.J., Suslow, T.V., Coaker, G.L., Leveau, J.H., 2012. Leaf microbiota in an agroecosystem: spatiotemporal variation in bacterial community composition on field-grown lettuce. The ISME Journal 6, 1812–1822. http:// dx.doi.org/10.1038/ismej.2012.32. Rosenblueth, M., Martinez-Romero, E., 2006. Bacterial endophytes and their interactions with hosts. Molecular Plant-Microbe Interactions 19, 827–837. Saitou, N., Nei, M., 1987. The neighbor-joining method-a new method for reconstructing phylogenetic trees. Molecular Biology and Evolution 4, 406–425. Semenov, A.V., van Overbeek, L., van Bruggen, A.H.C., 2009. Percolation and survival of Escherichia coli O157:H7 and Salmonella enterica serovar Typhimurium in soil amended with contaminated dairy manure or slurry. Applied and Environmental Microbiology 75, 3206–3215. Semenov, A.M., Kuprianov, A.A., van Bruggen, A.H., 2010. Transfer of enteric pathogens to successive habitats as part of microbial cycles. Microbial Ecology 60, 239–249. Solomon, E.B., Matthews, K.R., 2005. Use of fluorescent microspheres as a tool to investigate bacterial interactions with growing plants. Journal of Food Protection 68, 870–873. Solomon, E.B., Yaron, S., Matthews, K.R., 2002. Transmission of Escherichia coli O157:H7 from contaminated manure and irrigation water to lettuce plant tissue and its subsequent internalization. Applied and Environmental Microbiology 68, 397–400. Takeuchi, K., Frank, J.F., 2000. Penetration of Escherichia coli O157:H7 into lettuce tissues as affected by inoculum size and temperature and the effect of chlorine treatment on cell viability. Journal of Food Protection 63, 434–440. Takeuchi, K., Matute, C.M., Hassan, A.N., Frank, J.F., 2000. Comparison of the attachment of Escherichia coli O157:H7, Listeria monocytogenes, Salmonella typhimurium, and Pseudomonas fluorescens to lettuce leaves. Journal of Food Protection 63, 1433–1437. Zhang, G.D., Ma, L., Beuchat, L.R., Erickson, M.C., Phelan, V.H., Doyle, M.P., 2009a. Evaluation of treatments for elimination of foodborne pathogens on the surface of leaves and roots of lettuce (Lactuca sativa L.). Journal of Food Protection 72, 228–234. Zhang, G.D., Ma, L., Beuchat, L.R., Erickson, M.C., Phelan, V.H., Doyle, M.P., 2009b. Lack of internalization of Escherichia coli O157:H7 in lettuce (Lactuca sativa L.) after leaf surface and soil inoculation. Journal of Food Protection 72, 2028–2037.