Photometric technology is commonly used to measure two different phenomena, photons can be either absorbed or scattered by the sample being measured. When photons are absorbed by the sample absorbance is the measurement in question. When photons are scattered by the sample, turbidimetric measurement is involved. Understanding the difference between these two measurement types is especially important in assay development, data analysis and troubleshooting.
Thermo Scientific Multiskan SkyHigh microplate spectrophotometer and Thermo Scientific SkanIt software have distinct modes for absorbance and turbidimetric measurements to make it more clear to the users which type of measurement is being undertaken.
This note describes the difference between absorbance and turbidimetric modes and demonstrates two turbidimetric application examples, bacterial growth and endotoxin measurement performed with Multiskan SkyHigh spectrophotometer and SkanIt software.
Turbidimetry measures the presence of solid particles in a non-homogenous solution. Solid particles can be for example cells, precipitate, aggregation or any other solid mass that is present in the solution. The particles scatter the photons of the photometric beam and therefore a lower light intensity is detected by the instrument detector. This phenomenon is very different from absorbance measurement where photons are absorbed by a specific dye in the homogeneous solution. The basic difference between absorbance and turbidimetric scattering is shown in Figure 1.
Figure 1. Light intensity in the photodetector decreases in an absorbance assay because light is absorbed by the sample (A) and (B) also decreases in a turbidimetric assay because photons are scattered when they hit the solid particles.
The second remarkable difference comes from the fact that turbidimetric measurements are always performed with heterogeneous samples. Modern microplate photometers typically use rather narrow measurement beams because of the requirement to measure 384-well plates. Typically, the beams in microplate photometers beams have less than a millimeter beam diameter, and when that beam goes through the well of 96-well plate, only very small portion of the sample volume is covered by the measurement beam (Figure 2). At the same time, modern microplate photometers also use a very short sampling time, typically in milliseconds, which is not optimal for turbidimetric measurements as it also increases variation.
Figure 2. When a photometric beam with 0.7 mm diameter (used in the Thermo Scientific Varioskan LUX) passes through the sample in a 96-well plate (well bottom diameter typically 6.6 mm), the beam covers 1.1% of the total volume in the well (figure in scale).
Solid particles in liquid are understandably not in a static state but are freely floating throughout the liquid. Therefore, when a heterogeneous sample has only few solid particles, there is a low statistical probability that same number of particles would be within the beam in repeated measurements. For example, if there are 100 particles in the sample, the most probable number of particles in the measured beam area is 1 particle, but randomly there might be either zero or two particles, or even three. When sampling time is also in milliseconds, dynamic particle flow in the liquid does not change this situation remarkably in an individual measurement. As a result, with low OD values it is likely that assay precision will be markedly decreased and larger-than-normal CVs% will be observed in an absorbance assay.
Similarly, when number of the particles in the sample increases to very high numbers, then the photons scattered from the particle will hit another particle and re-scatter. This will cause some scattered photons to arrive at the detector following multiple scatterings. Therefore, the linear measurement range in turbidimetric measurements is always lower than in absorbance measurements. Microplate reader performance specifications are practically always given based on absorbance measurements; therefore one cannot expect those to be achieved when turbidimetric assays are in question.
The third very important difference between absorbance and turbidimetric measurement is that the instrument’s optical design affects the resulting OD value (Figure 3). Therefore, two different microplate photometers from different manufacturers cannot be expected to give the same OD values with the same sample. Two major features here are the size of the photodetector collection window, or collection slit if such is in use. The bigger the detector window, the more slightly scattered photons will be collected and therefore the lower the observed OD will be. Another major issue is the distance from the sample to the detector. If the detector is far away from the sample, even slightly scattered photons will arrive at the detector and if it is very close, then even strongly scattered photons will be collected by the detector. This distance is therefore influencing the resulting OD value very strongly. Since every manufacturer has a different optical design, these two parameters (distance and size of the detector) are never the same in two different instruments. These factors do not influence absorption measurements because beam shape remains intact in absorbance assays. Detector size or distance does not change amount of light arriving to the detector.
The fourth difference between absorbance and turbidimetric scattering assays is related to optical pathlength correction. Optical pathlength correction is a method where photometric pathlength is determined using water absorbance at 975 nm, and results are normalized for 10 mm optical pathlength. Solid particles in the sample will scatter also this 975 nm light, causing remarkable interference in pathlength determination. Therefore, reliably performing optical pathlength correction is not possible in turbidimetric scattering assays.
The main technical differences between absorbance and turbidimetric scattering assays are listed in Table 1.
Figure 3. The optical design of the microplate photometer heavily influences OD measurements. Detector collection window size and distance from the sample change the resulting OD values, and therefore different photometers will give different ODs with the same sample. The detected OD difference can easily be 20–40%.
|Absorbance measurement||Turbidimetric scattering measurement|
|The dye in the sample absorbs some of the light passing through the sample and therefore lower light intensity enters the detector||The solid particles in the sample scatter the photons and that cause the detected light intensity to be decreased|
|Always homogeneous sample||Always heterogeneous sample|
|Reported values use “Abs” as a unit||Reported values use “OD” as a unit|
|Specified linear measurement range typically >2 Abs units||Linear measurement range lower than specified for photometry|
|Specified accuracy and precision can be reached||Accuracy and precision are below specified values, especially at low ODs (bigger CV% can be expected)|
|Results of different photometers are very similar||Results of instruments can differ remarkably because of differences in optical design|
|Optical pathlength correction using water absorbance at 975 nm can be utilized||Optical pathlength correction is not possible|
Multiskan SkyHigh spectrophotometer offers two separate measurement modes with both SkanIt software 6.1 and instrument UI users: one for absorbance measurements and one for turbidimetric scattering measurements. From a technical and electronic point of view, these two measurement modes are 100% identical, the actual measurement will be performed by the instrument identically in both modes. The separation will help the user to clarify which kind of photometric assay is in question and offers the possibility of managing the results more clearly. For example, according to IUPAC (International Association of Pure and Applied Chemistry) recommendation, the results of absorbance measurements shall be referred with unit “Abs”, and results of turbidimetric scattering measurements as “OD”. This separation as well offers possibility to eliminate optical pathlength correction from those turbidimetric measurements where it is impossible to perform.
Figure 4 demonstrates the difference between results of a photometric and a turbidimetric assay. The deviation in the turbidimetric results shows that Multiskan Sky and SkyHigh photometers have very similar optical design and therefore the results are practically identical. Varioskan LUX and Multiskan FC photometers use a different type of optics and therefore differ remarkably from Sky and SkyHigh photometers, and those two also differ clearly from each other. In absorbance measurements all four microplate photometers give identical results.
Figure 4. Comparison of results of the microplate photometers in absorbance and turbidimetric measurements. The turbidity testing was performed with dilutions of a turbidity standard solution, (Formazin Standard Solution 4000 NTU, Sigma-Aldrich TURB4000) and the absorbance testing with a colorimetric Coomassie protein assay (Cat. No. 1856210).
Turbidimetric scattering is a very common method to analyze bacterial growth and perform antibiotic susceptibility testing as the growth of bacteria causes an increase in optical density of the sample. The scattering also depends on the particle size and form, but most bacteria have nearly the same absorbance per unit dry mass concentration. Thus, the light transmitted is inversely proportional to the number of bacteria. In bacterial assays the measurement wavelength needs to be selected on the range, where the absorbance by the biological material is minimal. Therefore, red wavelengths around 600–670 nm are typically used for these bacterial growth curve measurements.
In this example assay, 20 ml of Luria-Broth growth media was inoculated with 25 µl of an E. coli (ATCC 25922GFP) glycerol stock and incubated overnight at +37°C. The next morning the optical density of the bacterial culture was measured using a cuvette, and a fresh culture was started from the overnight culture. The fresh culture was prepared so that the OD620 at the beginning was 0.05 OD. The OD of the new culture was measured using a Thermo Scientific Nunc Edge plate (Cat. No. 167425) with 200 µl volume. One column of the plate was filled with media and used as blanks and the rest of the plate was filled with bacterial culture.
Multiskan SkyHigh spectrophotometer was pre-heated to 37°C. The plate was covered with lid, and OD values were read kinetically at 620 nm once every 10 min using background shaking. The shaking was set to be pulsed with 5s ON and 5s OFF and using medium shaking force. Measurements were taken for 18 hours. After the data were exported to Microsoft Excel from the instrument, user interface and the growth curves were created using blank-subtracted data. The measurement steps and the parameters of the SkanIt software session are shown in Figure 5. The average growth curve measured with the SkyHigh instrument user interface is shown on Figure 6. With SkanIt software, the kinetic data can be further processed by the kinetic and dose-response processors, enabling—for example—easy antibiotic susceptibility testing.
Figure 5. SkanIt software protocol settings for bacterial growth curve measurement.
Figure 6. Average growth curve of E. coli.
The other application example is a turbidimetric bacterial endotoxin assay. It is based on using the LAL reagent and coagulation. The reagent is an amebocyte extract from the horseshoe crab (Limulus polyphemus). It has been shown that when a gram-negative bacterium infects a horseshoe crab, the immune response is an intravascular clotting reaction. This coagulation results from a reaction between endotoxin and a clottable protein secreted by amebocytes. The same endotoxin clotting reaction is utilized in commercial LAL endotoxin assays. The formation of clot causes an increase in turbidity of the sample. This process can be measured in kinetic format on a microplate photometer. This example assay was performed with Pyrogent 5000 Kinetic Turbidimetric LAL Assay kit (Lonza, Cat. No. N383)
Multiskan SkyHigh was pre-heated to 37°C prior to the assay. The kinetic measurement was started by mixing a 100 µl aliquot of each sample (blank, standard or unknown) and 100 µl of the LAL reagent in a clear 96-well microplate well (Thermo Scientific Nunc MicroWell plate, (clear, 96-well, Cat. No. 167008). The kinetic measurement at 340 nm was started at constant temperature of 37°C. The formation of the endotoxin clot increases turbidity of the sample, and this process was followed kinetically. Each well was measured once in every 30 seconds for one hour. The plate layout and SkanIt session protocol and data analysis step three is shown in Figure 7.
Figure 7. The plate layout and SkanIt session steps for the turbidimetric endotoxin assay.
The clotting effect can be seen in the kinetic curves of the standards samples in Figure 8. Final endotoxin concentration results are calculated from the kinetic curves according to the kit instructions.
Figure 8. Kinetic result curves of the turbidimetric endotoxin assay. Clotting of the sample causes the increase in the turbidity. Even if the increase in turbidity is quite small, it is easily detected by the Multiskan SkyHigh spectrophotometer.
For Research Use Only. Not for use in diagnostic procedures.