> For the complete documentation index, see [llms.txt](https://help.sbtinstruments.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://help.sbtinstruments.com/mpd/cell-growth/optical-drawbacks/od-confounds-medium-optimization.md). # OD confounds medium optimization Optical density (OD) is commonly used as a proxy for cell concentration when optimizing growth media. The objective is typically to identify the medium composition and component concentrations that maximize cell concentration (carrying capacity). OD measurements are attractive because they are simple, inexpensive, and provide immediate results. While this makes them well suited for rapidly screening and shortlisting promising media formulations, caution should be exercised when using OD as the sole criterion for medium optimization. To illustrate this, we investigated the growth of *E. coli* in shake flasks containing a commercially available chemically defined fermentation medium supplemented with varying concentrations of glycerol as the carbon source. Growth was assessed using OD alongside direct measures of cell concentration to evaluate whether OD accurately reflected the optimal conditions for high cell concentrations. ## Take-home messages * OD measurements suggest a strong positive relationship between glycerol concentration and culture performance, implying that higher glycerol concentrations improve the growth medium. * Direct cell counts obtained with BactoBox® and plate counts contradict this simple trend, indicating that increasing glycerol concentration does not necessarily result in higher cell concentrations. * OD shows poor aggreement with viable cell concentrations determined by plate counts. * BactoBox® measurements show good agreement with plate counts, providing a more reliable estimate of cell concentration. * The bacterial size (CIZE) measured by BactoBox® offers a plausible explanation for why OD can be misleading, as changes in cell size influence light scattering independently of cell number. The complete investigation and supporting data is presented below. ## Study design The investigation was divided into two parts. First study was primarily done to investigate the effect of glycerol concentration when measured with OD and BactoBox®. In addition this study aimed to determine suitable incubation time to reach maximal cell concentrations. Second study was done to replicate the findings from the first study with more technical replicates. Moreover, the experiments aimed to investigate the effect of a broader range in glycerol concentration. {% columns %} {% column %} **1: Growth curve comparison of OD and BactoBox®.** * Three cultures with low, medium, and high glycerol concentration: 4, 7, and 10 mL/L. 4 and 10 mL/L represented the range recommended by the manufacturer of the medium, while the "medium" represented a center point. * Growth curves were tracked with single BactoBox® and OD measurements. {% endcolumn %} {% column %} **2: End-point study comparing OD, BactoBox®, and plate counts** * Incubation time: 12 hours * Ten glycerol concentrations: 2, 3, 4, 5, 6, 7, 8, 9, 10, and 20 mL/L. * BactoBox® and OD measurements were done in triplicates. * Plate counts were done by Compact dry TC plates in triplicates {% endcolumn %} {% endcolumns %} Bacto™ CD Supreme Fermentation Production Medium inoculated with *Escherichia coli* ATCC 8739 was used for all experiments. The inoculum was prepared in standard LB medium. Cultures were grown in 50 mL of medium in 250 mL baffled Erlenmeyer flasks fitted with vented caps, using an initial inoculum of 1 × 10⁷ cells/mL. ## Part 1: Growth curves for low, medium, and high glycerol The data below (Figure A) shows the relationship between the added carbon source (glycerol concentration) and the maximum OD reached during the growth curve. The near-perfect linear relationship (**R² ≈ 1**) suggests that increasing glycerol concentration results in higher OD values. Based on these data alone, one would conclude that higher glycerol concentrations consistently improve culture performance. In contrast to the OD measurements, BactoBox® analyses (Figure B) revealed virtually no relationship between glycerol concentration and direct cell concentration (R² = 0.0003). This indicates that the apparent improvement observed with OD does not reflect an increase in cell numbers. Instead, the results suggest that growth was limited by factors other than carbon availability, meaning that increasing the glycerol concentration alone does not necessarily enhance biomass production. The two methods yield conflicting results. But which one should you trust? Data from the peer-reviewed [performance qualification](https://www.sciencedirect.com/science/article/pii/S0167701225002003) demonstrate a strong correlation between BactoBox® measurements and plate counts for this *E. coli* strain. This suggests that the trends observed with OD may be confounded by factors other than cell concentration. In [#part-2-deeper-investigation-of-glycerol-concentration](#part-2-deeper-investigation-of-glycerol-concentration "mention") this hypothesis is further evaluated by comparing both methods with plate counts. Meanwhile, the CIZE metric (D) provides a plausible explanation for why OD can be misleading. At the time of maximum OD, cultures grown with 4 mL/L glycerol exhibited the smallest average cell size (CIZE, \~0.75 µm), despite having a cell concentration comparable to the 10 mL/L culture. The smaller cell size likely resulted in reduced light scattering and, consequently, a lower OD measurement. In contrast, the 7 mL/L culture had a lower cell concentration but a substantially larger average cell size (CIZE, \~1.0 µm), which likely contributed to its higher OD. Similarly, the 10 mL/L culture combined a high cell concentration with a larger average cell size (\~1.0 µm), resulting in the highest OD. These observations suggest that differences in cell size, rather than cell concentration alone, can substantially influence OD measurements.

Effect of glycerol-concentration investigated with OD and direct cell counts. E. coli cultivated in shake flasks with Bacto™ CD Supreme fermentation production medium. Glycerol was added as carbon source at a concentration of 4, 7, or 10 mL/L. Dashed lines in figure D correspond to CIZE (secondary axis)

## Part 2: Deeper investigation of glycerol concentration The first study was based on single measurement**s** at each sampling time point. To assess the reproducibility of the strong correlation between OD and glycerol concentration, a second study was performed using technical triplicates. Plate counts were included as an independent reference method for determining viable cell concentration. Because growth curve experiments are labor-intensive, the second study employed an end-point design. Based on the results of the first study, samples were collected after approximately 12 hours of incubation, corresponding to the time at which cultures reached their maximum cell concentration. ### Replication confirms initial findings The first study demonstrated a strong linear relationship between OD and glycerol concentration and lack thereof when investigated with BactoBox® measurements. Again, the OD data imply that increased glycerol concentration in the 2–10 mL/L range leads to better performance of the medium (R2 = 0.69). BactoBox® results contradict this finding (R2 = 0.12). The plate count data corroborate the BactoBox® findings with a poor relationship between glycerol concentrations and cell concentrations (R2 = 0.0013). The error bars on BactoBox® measurements are tighter than the CFU results demonstrating that BactoBox® is a more precise method than plate counts.

Repetition experiment to investigate the effect of different glycerol concentrations when measuring with OD600, BactoBox®, and plate counts. The linear regression is done for the data range 2–10 mL/L (squares). The data points at 20 mL/L were excluded (triangles). Error bars represent standard deviation of triplicate measurements.

### Statistical assessment In addition to the regression analyses, the optimal glycerol concentration was evaluated using Tukey's HSD post hoc test (family-wise α = 0.05). * **OD:** 10 mL/L is uniquely optimal within the tested glycerol concentrations. * **BactoBox®:** 10 and 20 mL/L perform similarly and both exceed the lower concentrations. * **CFU:** Differences are weak and only 10 mL/L is distinguishable from 6 mL/L. Overall, the statistical analyses reinforce the conclusion that the BactoBox® and plate count methods do not support the same interpretation as the OD measurements. Specifically, unlike OD, the direct cell-counting methods do not indicate that **10 mL/L** is uniquely superior within the tested glycerol range. ### OD demonstrates poor correlation with plate counts Plate counts are widely regarded as the reference method for viable cell concentration during growth medium optimization. As shown below (Figure A), OD and plate counts are poorly correlated (R² = 0.12), whereas BactoBox® measurements shows a good correlation with plate counts (R² = 0.83; Figure B). Together, these results demonstrate that OD was the confounding method in this study, whereas BactoBox® accurately reflected changes in cell concentration.

Correlation between plate counts and OD (A) and BactoBox® measurements (B). Dotted lines represent linear regression fits; error bars indicate the standard deviation of triplicate measurements.

## Conclusion Light scattering methods like OD may be useful for online measurements and quick screening with high throughput in e.g. 96 well plates. But OD can change for reasons other than cell concentration, especially when cells differ in size, shape, or growth state or when high background is present. This means that OD is often not a good proxy for max cell concentration. Don’t treat higher OD as automatically higher carrying capacity. A medium can give a higher OD because cells scatter more light, not because there are more cells. Instead of relying solely on OD and light scattering methods to develop the best medium, use the methods to narrow the field of candidates. This helps to eliminate clearly poor media recipes. Subsequently confirm and refine the shortlist with total direct counts provided by BactoBox® or other flow cytometric techniques. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://help.sbtinstruments.com/mpd/cell-growth/optical-drawbacks/od-confounds-medium-optimization.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.