The individual growth of Burkholderiaceae strains on Arabidopsis exudates
The ability of these four strains to use each of the four PRE as the sole carbon and energy source was first tested. To rule out a possible interference due to the use of sucrose during the germination of Arabidopsis, residual levels of sucrose were measured for each PRE (14d.PRE, 21d.PRE, 14d.N-PRE, and 21d.N-PRE) with average values of 0.078 ± 0.022; 0.076 ± 0.029; 0.022 ± 0.004; and 0.071 ± 0.050% p/v respectively (Table 1, first row), which represent values that showed no difference with this PRE. In agreement, tests performed in liquid cultures containing 0.1% p/v sucrose showed no growth for the four strains. Growth tests on A. thaliana root exudates showed that P. phytofirmans PsJN and C. pinatubonensis JMP134 reached statistically higher population levels than C. metallidurans CH34 and C. taiwanensis LMG19424, with each of the four PRE (Fig. 1). On average, P. phytofirmans PsJN and C. pinatubonensis JMP134 grew 5.4 times faster than strains CH34 and LMG19424 on N limited exudates, whereas P. phytofirmans.
Five different organic compounds were measured to explore possible differences between each plant root exudate. For more information on the techniques and protocols used in each case, refer to the Methods section. Each row shows the average of three technical replicas with their respective standard deviation. Comparisons between each exudate were made using Student’s t-test, and the significant differences are indicated with asterisks (*).
PsJN and C. pinatubonensis JMP134 grew 4.3 and 1.4 times faster than C. metallidurans CH34 and C. taiwanensis LMG19424 in 14d.PRE and 21d.PRE, respectively, excepting that C. metallidurans CH34 proliferated 1.4 times faster than C. taiwanensis LMG19424 in 21d.-PRE. This comparison showed that all strains started the stationary phase after 24–30 h, except for C. metallidurans CH34 growing on 14d.N-PRE, and C. taiwanensis LMG19424 growing on 14d.N-PRE and 21d.N-PRE, where stationary phases were only achieved later than 48 h of culture (Fig. 1). Stationary phases lasted more than 72 h for P. phytofirmans PsJN and C. pinatubonensis JMP134, or around 48 h for C. metallidurans CH34 and C. taiwanensis LMG19424. Maximum growth yields were higher on N-PRE than in PRE. Death phases were observed for P. phytofirmans PsJN in 14d.PRE and 21d.PRE, and C. pinatubonensis JMP134 in 14d.N-PRE and 21d.N-PRE, whereas C. metallidurans CH34 and C. taiwanensis LMG19424 showed a slower decline in bacterial cell numbers after 72 h in all conditions (Fig. 1).
The gross composition of these PRE was determined and is shown in Table 1. Five measurements were performed to determine residual sucrose levels, total phenolic, total carbohydrates, total protein, and total organic content. The measurements show no significant differences between the PRE on residual sucrose levels, total phenolic, and protein content. On the other hand, the total carbohydrate content found was significantly higher (double) in 14d.N-PRE from the other three PRE. Finally, total organic carbon contents were substantially lower (half) on 21d.PRE from the other three PRE.
Better together? Growth of combinations of Burkholderiaceae strains in Arabidopsis root exudates
Potential cooperation or competition interactions among bacteria were determined in growth cultures inoculated with a mixture of the four strains, starting (T0) at the same concentration (0.1 OD600nm). When these strains were grown together, the same growth pattern was observed with all tested PRE, with maximum yields ranging from 1.0–1.2 OD600nm and generation times of 7.1 h and 7.9 h on 14d.PRE and 21d.PRE, respectively, and of 9 h and 6.3 h on 14d.N-PRE and 21d.N-PRE, respectively (Fig. 1). The shapes of the co-culture growth curves were essentially like those observed for the individual growth curves of P. phytofirmans PsJN and C. pinatubonensis JMP134 in all PRE, except for the higher or lower growth levels transiently observed with P. phytofirmans PsJN grown on standard conditions PRE, and the death phases of C. pinatubonensis JMP134 occurring with N-PRE (Fig. 1). Bacterial abundances were determined for each strain growing in the 4-member combination at final growth times (120 h) (Table 2, Generation 1). Bacterial abundances of C. pinatubonensis JMP134 were 1–2 orders of magnitude higher than those of P. phytofirmans PsJN and C. taiwanensis LMG19424, except for the latter in 21d.NPRE. In contrast, absolute abundances of C. metallidurans CH34 were 1–3 orders of magnitude lower than the other three strains in all PRE. These results suggest that, at the end of the co-culture, C. pinatubonensis JMP134 was the main responsible for growth performance within the 4-member co-culture. This hypothesis was further studied using the k-means clustering algorithm, expecting C. pinatubonensis JMP134 to cluster with the 4-member co-cultured growth.
The growth pattern found for the 4-member co-culture and their individual growths were compared to analyze if the growth dynamics of the 4-member co-culture resembled that of any individual bacteria and, therefore, some of them dominate over the others in the co-culture (Fig. 2). It was observed that the co-culture grouped with P. phytofirmans PsJN and C. pinatubonensis JMP134 in all PRE (Fig. 2A-C), except 21d.N-PRE, where C. pinatubonensis JMP134 grouped only with two co-cultures replicates (co-culture_8 and co-culture_5), and P. phytofirmans PsJN grouped with the remaining six replicates (Fig. 2D). On the other hand, significant changes among PRE were explored. The results show that only two clusters were determined for 14d.PRE and 21d.PRE (Fig. 2A&B), the first composed of P. phytofirmans PsJN, C. pinatubonensis JMP134, and the co-culture, and the second formed by C. metallidurans CH34 and C. taiwanensis LMG19424. On the other hand, the cluster determined for PREs obtained from plants under N-limiting conditions was less homogeneous. On the one hand, in 14d.N-PRE (co-culture_8 was considered as an outsider [Fig. 2C]), three clusters were determined, while 21d.N-PRE displayed four significant clusters (Fig. 2D). These results demonstrated a clear effect of the type of exudates with both individual strains and co-culture. Also, they corroborate the previous observation that growth curves in the co-culture were mainly influenced by P. phytofirmans PsJN and C. pinatubonensis JMP134.
It is worth mentioning that no positive or negative effects between pair combinations grown under two standard laboratory conditions were found. LB and R2A plate cross strike tests revealed no growth inhibition halos. In addition, growth and survival tests performed in spent media (i.e., liquid culture media after growth of one of these four strains) in LB and 5 mM succinate Dorn minimal medium showed no decrease/increase in survival (measured as CFU/mL) or growth (OD600nm) after 48 h of incubation. These results indicate that no inhibitory compounds nor growth-enhancing molecules were produced upon the development of the first strain.
Exploring microbial interactions through combinatorial co-culturing
To further explore interactions between these strains that would explain the final abundances of the 4-member co-culture (Table 2, Generation 1) and the different aggrupation found in the cluster analysis (Fig. 2), combinatorial co-culture growth tests were carried out to determine viable cell counts. Since 14d.N-PRE and 21d.N-PRE absolute bacterial abundances were similar (Table 2, Generation 1), co-culturing tests were carried out only with 14d.PRE, 21d.PRE and 21d.N-PRE. Individual growth levels were compared with those determined in pairs, trios, and the 4-member combinations (Additional File 1), and the corresponding percent variations in viable cell numbers were calculated (Fig. 3). PRE heavily modified cell numbers of each strain growing in combinations. Decreases in the abundances were more frequent than increases (Fig. 3), indicating that inhibitory interactions predominated. Percent variation increases were observed for P. phytofirmans PsJN (6 to 26%) co-cultures grown on 14d.PRE, and to a lesser degree for C. pinatubonensis JMP134 (7–9%) and C. taiwanensis LMG19424 (4–13%), with no essential differences if the co-culture consisted of pairs, trios, or the full quartet, except in two cases (C. taiwanensis LMG19424 when is paired with C. pinatubonensis JMP134, and JMP134 on the quartet arrangement) where C. taiwanensis LMG19424 was part of the co-culture (Fig. 3). Although C. metallidurans CH34 had decreased growth on any combination (1–15%), this effect was more significant on co-cultures with strains LMG19424 and JMP134. The 21d.PRE negatively affected the growth of C. pinatubonensis JMP134 (14–24%), C. metallidurans CH34 (7–18%), and C. taiwanensis LMG19424 (7–12%), except for minor increases (1–2%) for P. phytofirmans PsJN, but not in the presence of C. pinatubonensis JMP134, which may be related to the better performance of P. phytofirmans PsJN in this exudate. In contrast with 21d.PRE, the 21d.N-PRE, consistently decreased growth for all strains when tested in co-cultures (Fig. 3), with C. taiwanensis LMG19424 being the most affected (3–52%), C. metallidurans CH34 and C. pinatubonensis JMP134 decreasing between 3 and 10%, and P. phytofirmans PsJN showed decreases (2–13%), or slight increases (1–2%).
Sequential transfer dynamics of four-member co-culture
To study potential fitness changes in time, the abundances of these Burkholderiaceae strains were determined after six sequential transfers, i.e., seven generations (Fig. 4). Decreases in maximal growth were observed after the first sequential transfer (Fig. 4A, C&D), except for 21d.PRE after the 4th generation, where an increase in growth was detected (Fig. 4B). Also, the shapes of the growth curves changed with longer lag phases: 12.5-, 15.8-, 19.2-, and 10-fold average increases for 14d.PRE, 21d.PRE, 14d.N-PRE, and 21d.NPRE, respectively, and slower generation times: 1.2-, 2.7-, 3.0-, and 1.5-fold average increases for 14d.PRE, 21d.PRE, 14d.N-PRE and 21d.N-PRE, respectively. There were, however, some differences between 14d.PRE, 14d.N-PRE, and 21d.N-PRE. For the former, changes were steadily observed through the initial generations, e.g., maximal 14d.PRE growth yields of 0.61, 0.48, 0.2, and 0.15 OD600nm, were detected (Fig. 4A), whereas for the latter, a sharp decrease was detected soon after the first transfer, e.g., maximal 14d.N-PRE growth yields of 0.9, 0.15, 0.19, 0.16, 0.07, and 0.21 OD600nm, were observed (Fig. 4C). For 14d.PRE and 14d.N-PRE from the 4th generation onward, the 4-member co-culture never recovered, and no growth could be detected (Fig. 4A&C). On the other hand, with 21d.PRE after the 4th generation, the 4-member co-culture has a non-stable behavior (Fig. 4B). The 5th generation showed a 12 h lag phase followed by a log phase with a generation time of 3.53 h, reaching a maximum OD600nm value of 0.22. The 6th generation showed a long lag phase (56 h) followed by a short log phase that reached a maximum OD600nm of 0.16 at 72 h. Finally, the 7th generation showed a 10 h lag phase followed by a log phase of 8.4 h and a maximum OD600nm of 0.8 at 72 h (Fig. 4B). On the other hand, the 4th member co-culture showed growth in all generations in N-PRE, regardless of the age of the plant from which the exudate was collected (Fig. 4 C&D). These results showed that growth on N-PRE decreased steadily over time (e.g., maximal growth on 21d.N-PRE from 4th generation: 0.08, 0.07, 0.04, 0.02).
To analyze if the abundances of the members of this co-culture changed over time, viable cell counts were determined at the end of each culture (Table 2). Results for 14d.PRE showed complete loss of viable cells since the 4th generation, except for C. metallidurans CH34 which disappeared in the 3rd generation. Individual abundances were determined after growth on 21d.PRE showing that the four strains decreased in viable cell numbers after sequential transfers. P. phytofirmans PsJN and C. pinatubonensis JMP134 remained at significant levels even after the sixth transfer, with C. pinatubonensis JMP134 always showing higher abundance levels (Table 2). In contrast, C. metallidurans CH34 and C. taiwanensis LMG19424 completely disappeared after the 5th and 6th generation, respectively. A different pattern was observed with both N-PRE. The four strains remained viable at significant levels even after the sixth transfer, although C. metallidurans CH34 exhibited levels two orders of magnitude lower than the other strains, with C. pinatubonensis JMP134 showing at least one order of magnitude higher levels than the other strains. However, viable cell counts diminished after sequential transfers with the four strains, especially in 21d.N-PRE. Taken together, these results indicate that the inability of this 4-member combination to sustain growth on PRE depends on the type of PRE and the bacterial interactions.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.