Increased growth temperature and vitamin B12 supplementation reduces the lag time for rapid pathogen identification in BHI agar and blood cultures

Bacterial strains

The study focused on seven different bacterial strains covering four different bacterial species. Clinical bacterial strains E. coli NCTC13441 and S. aureus NCTC8325 were obtained from the National Collection of Type Cultures (NCTC), while S. aureus CCUG35600 and E. coli CCUG17620 were obtained from the Culture Collection University of Gothenburg (CCUG). Three non-clinical wild-type strains, including E. coli HB101 K2, Pseudomonas aeruginosa INN, and Bacillus subtilis INN, were taken from the internal bacterial collection of Inland Norway University (INN). Additional details are available in Table 1.

Chemicals

All reagents were obtained from standard lab suppliers. Bacteriological agar and Brain Heart Infusion (BHI) broth for the media preparation were the products of HiMedia Laboratories (Mumbai, India; lot number: BCBH4004V), with BHI consisting of (g/L) calf brain infusion from 200 g (12.5), beef heart infusion from 250 g (5), peptone (10), sodium chloride (5), disodium chloride D (+) glucose (2), disodium hydrogen phosphate (2.5) at pH 7.4. Vitamin B12 or cyanocobalamin was from ThermoFisher, (Massachusetts, USA; number A14894). In addition, we also tested three clinical isolates using a trace metal solution (details in Supplementary Table 1, Extended data [Ahmad 2023]).

Pre-culture and growth procedures

BHI agar and liquid culture media were used to maintain cultures and to examine growth curves. Bacterial samples stored at -80°C were thawed gently at room temperature, inoculated to BHI broth, and incubated overnight at 37°C. BHI agar plates were prepared by adding 1.5% bacteriological agar powder to BHI broth. An overnight inoculum was streaked out, and plates were incubated at 37°C for 24 h. Single colonies were selected and inoculated into fresh BHI broth. The inoculated BHI broth was incubated at 37°C overnight and served as inoculum for culture experiments (See Supplementary figure S1 for an overview of the complete process, Extended data). Experimental liquid cultures were cultivated in shake flasks (500 mL broth in 2L Erlenmeyer bottles). Cultures within the same experimental design were inoculated from the same bacterial preculture suspension to ensure identical starting conditions.

All cultures in BHI broth were incubated at 37°C or 42°C under shaking conditions for up to 6 h, whereas blood cultures were cultivated for 12 h. During incubation, the optical density (OD) was measured at 600 nm every 20-30 minutes, starting from the time of inoculation (0 min) until the bacterial growth reached the stationary phase (approximately 4-5 h). Culture aliquots were immediately cooled and kept on ice until measurement. The OD 600 was plotted versus time (min) to obtain the growth curves.

As a practical response value to estimate the effectiveness of treatments, an endpoint of the lag time (t_lag) was defined, the t_lag being the time (min) from inoculation until the OD 600 value reached a net increase of 0.5. The net OD increase was obtained by subtracting the OD at t=0 from all OD values. The t_lag was obtained by extrapolating at OD 0.5 from the zero-adjusted growth curve. Although this provisional end point is beyond the actual end of lag (Bertrand, 2019), an extended OD cut-off value was selected to reduce the estimation error when growth curves go from an undefined shallow rise to a steady exponential growth (Figure 1).

Figure 1. Graphic description of the determination of t_lag from the growth curves used in this study.

The 0.5-intercept represents the net increase in OD 600, i.e., the value subtracted from the optical density (OD) at time zero.

The doubling time (t_d) for bacteria was calculated based on the steepest rise in the exponential phase of the growth curve. The OD (log) and time interval in the exponential phase were selected at two points, which showed a specific growth rate (μ). The following formula was used to calculate the bacterial doubling time.

In the formula above, 0.693 is constant, and “μ” is the specific growth rate, which is calculated by using the following equation:

where;

Ln is the natural log

OD₂ is the OD at the endpoint of the exponential phase

OD₁ is the OD at the starting point of the exponential phase

t₂ is the time corresponding to OD₂

t₁ is the time corresponding to OD₁

Design of experiments for screening of growth variables

A set of growth screening experiments examined the ability of selected variables to cause a reduced lag time. A structured two-level factorial design (Design of Experiments, DOE) was used to observe the effect of incubation temperatures (37°C and 42°C), low (0.05%), and high (0.5%) inoculum, vitamin B12 (±50 μM), and trace metal (present y/n) supplements. The design was tested on the wild-type strains (3) and the clinical isolates (4) but with a selection of two or three variables or factors. Table 2 shows the coded and decoded values of the factors tested (X₁, X₂, X₃) at a low (-1) and high (+1) level. Thus, either four (2²) or eight (2³) growth experiments were performed for each strain. For a detailed overview of conditions for the individual strains, Table 1 is provided as well as in Supplementary Tables 2-4 (Extended data).

The estimated t_lag values from DOE were presented in interaction effect plots, illustrating the impact of a variable (e.g., temperature) with connected experiments on another variable (e.g., vitamin B12 supplement). The main effects of tested variables are shown numerically in descending order in bar charts. The bars are the absolute effect values computed as twice the regression coefficients from fitted DOE models. They show the change in t_lag when a variable goes from a low to a high level when other factors are kept at their average.

Blood culturing

Clinical strain E. coli CCUG17620 was spiked with blood to mimic the sepsis infection. The blood was spiked with 40 CFU/mL from an overnight plate culture of the clinical strain. The lower CFU was used as an inoculum because, in the case of sepsis, the CFU/mL ranges typically from 10-100 (Doualeh et al., 2022). E. coli was grown overnight on a BHI agar plate and then suspended in saline to measure the OD. After measuring OD, the suspensions were plated on agar plates to calculate the CFU at different OD values. After optimizing the relation of OD with CFU counts, the desired concentration in the range of 10-100 CFU/mL was used for spiking with human blood. Fresh human blood obtained from healthy donors at Inland Norway University was used for this experiment. Human blood spiked with E. coli was mixed with BHI broth medium and incubated at the two study temperatures, i.e., 37 and 42°C. CFU/mL was calculated every 3 h for 12 h by plating out the samples on BHI agar plates. The CFU/mL was calculated using the formula: number of colonies × dilution factor/volume of culture plated.

DNA extraction and PCR from blood cultures

Bacterial DNA from blood culture samples was extracted using QIAamp BiOstic Bacteremia DNA Kit from Qiagen (Germany). The concentration and purity of isolated DNA were determined using the Qubit system with a dsDNA HS assay kit (Thermo Fisher Scientific, USA) and the NanoDrop spectrophotometer. PCR was performed to confirm the presence of E. coli CCUG17620 DNA and human DNA in samples taken at 3 h, 6 h, 9 h, and 12 h of incubation at 37°C and 42°C. The primers used for identifying E. coli were uspA-F CCGATACGCTGCCAATCAGT and uspA-R ACGCAGACCGTAGGCCAGAT, and human DNA was β-actin (F-CGGCCTTGGAGTGTGTATTAAGTA and R-TGCAAAGAACACGGCTAAGTGT) (Hasan et al., 2016). The PCR reaction was set up in a total volume of 20 μL per sample containing 4 μL HOT FIREPol MultiPlex Mix Ready to Load master mix from Solis Biodyne (Estonia), 0.5 μL of 10 μM forward and reverse primers, 8 ng/μL of template DNA and nuclease-free water. The thermal cycling conditions for PCR were initial denaturation at 95°C for 12 min followed by 35 cycles of 95°C for 25 s, 61°C for 50 s and 72°C for 1 min, and a final extension of 72°C for 7 min. The PCR product was run on 1% agarose gel for 45 min at 100V and then visualized on the Gel doc system to determine the presence or absence of our desired bands.

Statistical analysis

The MODDE 13.0 Pro (Umetrics/Sartorius Stedim Data Analytics AB, Sweden) was used to generate the design matrix, regression coefficients, and statistical analysis. The significance testing (p < 0.05) of the DOE screening was performed with ANOVA. All the growth curve graphs are generated using GraphPad Prism 9.4.1.

Higher incubation temperature, along with vitamin B12 supplementation, reduces the lag time for bacterial growth

Escherichia coli

E. coli HB101 K2

In the absence of vitamin B12 supplementation, the t_lag for bacterial growth was 265 minutes and 255 minutes at 37°C and 42°C, respectively, when an inoculum size of 0.05% was used for culturing (Figure 2A). When using a higher inoculum of 0.5% in the absence of vitamin B12 supplement, the t_lag recorded was 220 minutes and 145 minutes at 37°C and 42°C, respectively (Figure 2A). In the presence of vitamin B12 supplement and using an inoculum size of 0.05%, the bacterial t_lag was 315 minutes (37°C) and 160 minutes (42°C). Similarly, when the inoculum size was increased to 0.5%, with vitamin B12 supplementation, the t_lag was 200 minutes (37°C) and 190 minutes (42°C). The shortest t_lag was recorded to be 145 minutes, which was observed using an inoculum size of 0.5% in the absence of vitamin B12 supplement and incubation at 42°C (Figure 2 and Supplementary Table 2 [Extended data]).

Figure 2. Results from E. coli HB101 K2 under different growth conditions, including inoculum size, temperature, and vitamin B12 supplementation.

(A) Growth curves of E. coli HB101 K2 were obtained in three-factorial designs; at 37°C and 42°C, inoculum size 0.05% and 0.5 %, with and without vitamin B12. The dotted line indicates OD 0.5 for estimating the t_lag from curves. (B) Effect plot of E. coli HB101 K2 in minutes (t_lag) to reach optical density 0.5.

Using an inoculum size of 0.05% and a vitamin B12 supplement, we observed 95 minutes of reduction in the lag time at 42°C compared to 37°C (Figure 2B). In contrast, when the inoculum size was increased to 0.5% along with vitamin B12 supplementation, the t_lag was reduced by 10 minutes at 42°C. Based on the effect plot, it is reported that the sample with no vitamin B12 supplement and 0.5% inoculum size showed the shortest t_lag. But when the inoculum size was decreased to 0.05%, we observed that the sample incubated at 42°C in the presence of vitamin B12 supplement showed the shortest t_lag (Figure 2B). The reduction in t_lag was between 4% and 49% for the above mentioned eight tested conditions.

E. coli CCUG17620 and E. coli NCTC13441

Based on the results from the laboratory strain, we tested the growth of two E. coli clinical isolates at 37 and 42°C. E. coli CCUG17620 took 170 minutes when incubated at 42°C and 193 minutes at 37°C to reach OD 0.5 without adding vitamin B12. Similarly, E. coli NCTC13441 required 126 and 167 minutes to reach OD 0.5 at 42°C and 37°C, respectively (Figure 3A and Supplementary Table 4 [Extended data]). For the two clinical isolates, E. coli CCUG17620 and E. coli NCTC13441, the t_lag was less at 42°C compared to 37°C.

Figure 3. Results for the clinical strains E. coli CCUG17620, E. coli NCTC13441, and S. aureus CCUG35600 under different growth conditions, including temperature, and vitamin B12 supplementation.

(A) Growth curves of clinical bacterial isolates at 37°C and 42°C (E. coli CCUG17620, E. coli NCTC13441), or with and without vitamin B12 supplement (S. aureus CCUG35600 at 42°C). The dotted line indicates OD 0.5 for estimating the t_lag from curves. (B) Effect plot of clinical samples in minutes (t_lag) to reach optical density 0.5.

Overall, the effect plot of E. coli CCUG17620 and E. coli NCTC13441 at inoculum size 0.5% indicated that when the incubation temperature increased (42°C instead of 37°C), the lag time decreased by 15-25% in the tested strains (Figure 3B).

Staphylococcus aureus

S. aureus NCTC8325

In the case of the clinical isolate S. aureus NCTC8325 with no vitamin B12 supplements and 0.05% inoculum size, the time to reach OD 0.5 was 285 minutes and 210 minutes at 37°C and 42°C, respectively (Figure 4A and Supplementary Table 2). But in the presence of vitamin B12 supplementation and lower inoculum size of 0.05%, the t_lag was 190 minutes (37°C) and 165 minutes (42°C). When the inoculum size was increased to 0.05%, the t_lag decreased significantly at 42°C (135 minutes) in comparison to 37°C (215 minutes) in the presence of vitamin B12 supplementation (Figure 4A and Supplementary Table 2). We can conclude that incubation at 42°C, along with vitamin B12 supplement, reduces the t_lag in S. aureus NCTC8325.

Figure 4. Results for S. aureus NCTC8325 under different growth conditions, including inoculum size, temperature, and vitamin B12 supplementation.

(A) Growth curves of S. aureus NCTC8325 were obtained in three-factorial designs; at 37°C and 42°C inoculum sizes 0.05% and 0.5 %, with and without vitamin B12. The dotted line indicates OD 0.5 for estimating the t_lag from curves. (B) Effect plot of S. aureus NCTC8325 in minutes (t_lag) to reach optical density 0.5.

The effect plot of S. aureus NCTC8325 at 0.5% inoculum shows a reduction of 121 minutes in t_lag at 42°C incubation in the presence of vitamin B12 supplement. At the same time, a reduction of 45 minutes in t_lag was observed at 0.05% inoculum and 42°C incubation in the presence of vitamin B12 supplement (Figure 4B). The reduction in t_lag was between 13% and 37% for the eight tested conditions.

S. aureus CCUG35600

Based on the above results, we tested the growth of another clinical strain of S. aureus. The effect plot of the clinical isolate S. aureus CCUG35600 at inoculum size 0.5% at 42°C co-cultured with vitamin B12 supplement showed a decline in lag time compared to without supplementation. Overall, the shortest t_lag time was observed following the co-culture of bacteria with vitamin B12 supplement (Figure 3A and B). So, the reduction in t_lag was 20%.

Pseudomonas aeruginosa

P. aeruginosa showed a significant reduction in the t_lag when vitamin B12 supplements were added along with 42°C incubation. It reached OD 0.5 in 185 minutes when vitamin B12 supplement was added along with 42°C incubation. On the other hand, 235 minutes was required with no vitamin B12 supplement and 37°C incubation (Figure 5A).

Figure 5. Results for Pseudomonas aeruginosa and Bacillus subtilis under different growth conditions, including temperature and vitamin B12 supplementation.

(A) Growth curves of P. aeruginosa and B. subtilis were obtained in two-factorial designs, at 37°C and 42°C, with and without vitamin B12. The dotted line indicates OD 0.5 for estimating the t_lag from curves. (B) Effect plot of P. aeruginosa and B. subtilis in minutes (t_lag) to reach optical density 0.5.

The effect plot showed a reduction of 30 minutes in t_lag when bacteria were cultured at 42°C in the presence of vitamin B12 supplementation (Figure 5B). The decline in t_lag was between 9% and 20% for the four tested conditions.

Bacillus subtilis

B. subtilis also showed the same pattern as other bacteria, with a t_lag of 155 minutes with vitamin B12 supplement and 42°C incubation. On the other hand, the t_lag was 210 minutes and 215 minutes to reach OD 0.5 when incubated at 37°C with and without vitamin B12 supplement, respectively (Figure 5A and Supplementary Table 3 [Extended data]). The effect plot showed a reduction of 45 minutes in t_lag when bacteria were co-cultured at 42°C in the presence of vitamin B12 (Figure 5B). The reduction in t_lag was between 5% and 28% for the four tested conditions.

The main effects on t_lag from the variables tested in three- (E. coli and S. aureus) and two-factor (P. aeruginosa and B. subtilis) experimental designs are shown in Figure 6. We tested the time to reach OD 0.5 for S. aureus, E. coli, P. aeruginosa, and B. subtilis and noticed that incubating the bacteria at 42°C reduces the lag time. However, the clinical isolates (E. coli CCUG17620, E. coli NCTC13441, and S. aureus CCUG35600), which were tested for trace metal solution complement, adversely affected bacterial growth by prolonging the lag time (details in Extended data).

Figure 6. Main effects on t_lag from the variables tested in three- (E. coli and S. aureus) and two-factor (P. aeruginosa and B. subtilis) experimental designs.

Bars represent the change in t_lag (min) when a factor varies from its low to a high level (cfr. Table 1) when all other factors are kept at their averages. Calculated effects are given with 95% confidence. Other statistical features are given at the bottom.

Bacterial doubling time

The doubling time of all the tested isolates falls between 12 – 55 minutes. The median of the calculated t_d is 30.6 minutes (n=36), meaning the value closest to the midpoint of the population, i.e., the most ‘normal’ value. Although the difference in t_d was insignificant, we still noticed that at 42°C, t_d is lower (bacteria grow faster) than at 37°C. The difference in t_d at both temperatures was more pronounced in E. coli isolates (E. coli HB101 K2, E. coli NCTC13441, and E. coli CCUG17620) compared to S. aureus isolates. We did not see any effect of temperature on t_d in P. aeruginosa and B. subtilis isolates (Table 1 and Supplementary figure 3 [Extended data]).

Higher incubation temperature increases bacterial growth in blood cultures

We also observed that higher incubation temperature could increase bacterial growth in blood cultures as in pure bacterial isolates. The CFU of the blood samples, which were spiked with E. coli CCUG17620 and incubated at the two study temperatures, 37°C and 42°C, showed that bacteria could grow much faster in the early stages of incubation at 42°C compared to 37°C. When we calculated the CFU after 3 h and 6 h of incubation, we could see that the CFU/mL of bacteria were three times higher at 42°C than at 37°C. This higher growth of bacteria in the initial hours of incubation is critical for identifying the causative pathogen. This information is valuable for infection diagnostic purposes. Further incubation beyond 6 h, narrowed the gap between CFU/mL at 37°C and 42°C. At 12 h of incubation, we got the same CFU at both temperatures (Figure 7).

Figure 7. Increased viable counts (CFU/mL) of E. coli CCUG17620 spiked in blood cultures cultivated at 37 and 42°C.

PCR results show higher purity of DNA at an increased growth temperature

We performed PCR to confirm the presence of bacterial DNA versus human DNA at different incubation time points, as described in Methods. The PCR product run on 1% agarose gel was visualized for the presence of DNA and compared with a 100 bp DNA ladder, which was also run on the gel. We could see the DNA bands for the E. coli CCUG 17620 sample even after 3 h incubation. Although the CFU/mL after 3 h of incubation was low, i.e., 40 at 37°C and 120 at 42°C, it was still enough bacterial DNA to be amplified through PCR (Figure 8). For the samples which were incubated for a longer time and had greater CFU, the band signals for E. coli DNA appeared stronger in comparison to the human DNA (Figures 8 and 9). We can see from the PCR results that we obtained amplification at both the incubation temperatures (37°C and 42°C), but the DNA band strength and purity were higher at 42°C. The DNA concentration was also higher at increased incubation temperature (42°C) compared to 37°C. The DNA concentration of the samples after 3 h of incubation was 52.4 and 34.3 μg/mL at 42°C and 37°C, respectively. Similarly, we also observed an increased DNA concentration at 6 h (110 μg/mL at 42°C and 82 μg/mL at 37°C) and 9 h (134.4 μg/mL at 42°C and 119.2 μg/mL at 37°C) of incubation in samples which were incubated at a higher temperature. This higher concentration of DNA is beneficial for nanopore sequencing, which can be used to identify the bacteria in real time.

Figure 8. Blood cultures PCR (E. coli CCUG17620).

Lane 1: Negative control (No DNA), Lane 2: 12 h sample (42°C), Lane 3: 12 h sample (37°C), Lane 4: 9 h sample (42°C), Lane 5: 9 h sample (37°C), Lane 6: 6 h sample (42°C), Lane 7: 6 h sample (37°C), Lane 8: 3 h sample (42°C), Lane 9: 3 h sample (37°C), Lane 10: 100 bp DNA ladder.

Figure 9. Blood cultures PCR (Human).

Lane 1: Negative control (No DNA), Lane 2: 12 h sample (42°C), Lane 3: 12 h sample (37°C), Lane 4: 9 h sample (42°C), Lane 5: 9 h sample (37°C), Lane 6: 6 h sample (42°C), Lane 7: 6 h sample (37°C), Lane 8: 3 h sample (42°C), Lane 9: 3 h sample (37°C), Lane 10: 100 bp DNA ladder.