3.3.3 Semi-quantitative
Comparison
of Mycobacterial Lipid
Families Present in Studied
Samples
The area under the extracted-ion chromatogram of a given lipid
species correlates with its abundance. However, translating this area
in absolute quantity requires performing dose response curves
using standards, which is still challenging for mycobacterial lipids
due to the poor availability of standards. Also, differences in chromatograms areas between lipid classes cannot be translated into
differences in absolute quantity without determining the response
factor of each lipid class. Nonetheless, changes in the area under the
chromatogram for a given lipid across samples of the same nature
reflect changes in absolute quantity. Hence, the comparison of the
lipid profiles obtained for different samples of the same nature
allows highlighting changes in the relative proportion of the different lipid families without knowing their absolute value.
Normalization of the experiment is crucial to perform such
semi-quantitative comparative analysis. Since the use of detergent
in mycobacteria culture is not recommended, strong bacterial
aggregation is observed and optical density of the cultures cannot
be used for the normalization of the comparative analysis of different mycobacterial strains (see Note 4). Instead, comparative analysis
can be normalized based on the weight of extracted bacteria pellet
or on the amount of injected lipid extract (assuming that the same
amount of lipids is extracted per bacillus for different strains) (see
Note 8).
These considerations being said, the comparative analyses of
samples should be performed as follows:
1. Add an internal standard(s) (see Note 5) to studied samples
before lipid extraction to control the quality of the extraction
process and normalization.
2. Perform HPLC-MS analysis as described above by injecting the
amount of lipid extract according to the chosen normalization
method (see Notes 8 and 18) and to the previously determined
range of linear response of the studied lipid families (see Note
16).
3. Extract for each dataset the ion chromatograms at the expected
m/z for representative or all acyl-forms of the lipid families (see
Note 14), with a mass tolerance within the mass spectrometer
mass accuracy (1–10 ppm), and integrate signals under the
chromatogram curves.
4 Notes
1. Handle all organic solvents under the hood. Chloroform can
be replaced by dichloromethane, which is less toxic.
2. This approach was initially developed using an Agilent Technologies 6520 Accurate-Mass Q-Tof and a 1200 series HPLC
570
Emilie Layre
Comparison
of Mycobacterial Lipid
Families Present in Studied
Samples
The area under the extracted-ion chromatogram of a given lipid
species correlates with its abundance. However, translating this area
in absolute quantity requires performing dose response curves
using standards, which is still challenging for mycobacterial lipids
due to the poor availability of standards. Also, differences in chromatograms areas between lipid classes cannot be translated into
differences in absolute quantity without determining the response
factor of each lipid class. Nonetheless, changes in the area under the
chromatogram for a given lipid across samples of the same nature
reflect changes in absolute quantity. Hence, the comparison of the
lipid profiles obtained for different samples of the same nature
allows highlighting changes in the relative proportion of the different lipid families without knowing their absolute value.
Normalization of the experiment is crucial to perform such
semi-quantitative comparative analysis. Since the use of detergent
in mycobacteria culture is not recommended, strong bacterial
aggregation is observed and optical density of the cultures cannot
be used for the normalization of the comparative analysis of different mycobacterial strains (see Note 4). Instead, comparative analysis
can be normalized based on the weight of extracted bacteria pellet
or on the amount of injected lipid extract (assuming that the same
amount of lipids is extracted per bacillus for different strains) (see
Note 8).
These considerations being said, the comparative analyses of
samples should be performed as follows:
1. Add an internal standard(s) (see Note 5) to studied samples
before lipid extraction to control the quality of the extraction
process and normalization.
2. Perform HPLC-MS analysis as described above by injecting the
amount of lipid extract according to the chosen normalization
method (see Notes 8 and 18) and to the previously determined
range of linear response of the studied lipid families (see Note
16).
3. Extract for each dataset the ion chromatograms at the expected
m/z for representative or all acyl-forms of the lipid families (see
Note 14), with a mass tolerance within the mass spectrometer
mass accuracy (1–10 ppm), and integrate signals under the
chromatogram curves.
4 Notes
1. Handle all organic solvents under the hood. Chloroform can
be replaced by dichloromethane, which is less toxic.
2. This approach was initially developed using an Agilent Technologies 6520 Accurate-Mass Q-Tof and a 1200 series HPLC
570
Emilie Layre