Patent Application
20080108144
The present invention generally relates to improvements in the area of sample analysis particularly the analysis of samples that contain polypeptides. The invention provides improved sample preparation techniques as well as improved methods of analysis of samples. The improved techniques find particular application in the area of detecting the presence of polypeptides and polypeptide variants within a material. In a particularly preferred embodiment the invention relates to the detection of polypeptide variants by MALDI ToF mass spectrometry. The detection of polypeptide variants is of importance as the presence of polypeptide variants may be indicative of the presence of genetic abnormalities and/or the presence of other undesirable medical conditions.
The ability to accurately analyse materials for the presence of components such as polypeptides is an area growing in importance since the completion of the human genome project. Now that the genetic sequences have been provided it is increasingly important to be able to determine the components of materials in order to provide further information of interest on the material or the organism from which it was sourced. There is therefore an increasing need to provide improved methods of sample analysis of materials that contain components such as polypeptides. This analysis can provide information on the identity of polypeptides and polypeptide variants within the material. This information can be helpful in the diagnosis of certain medical conditions or the characterisation of mutant proteins.
Polypeptides are encoded by DNA and play important roles in most biological functions within organisms. The function performed by a polypeptide is determined by its structure, wherein the specific structure of the polypeptide allows specific interactions to occur with other molecules. The structure of a polypeptide is determined by the interaction of the amino acid side chains of the polypeptide with each other. Thus the overall structure, and hence the specificity, of a polypeptide is ultimately determined by its amino acid sequence.
As the amino acid sequence of a polypeptide is determined by the nucleotide sequence of its corresponding gene, mutations in genes can manifest themselves as variant polypeptides. Variant polypeptides may have altered function and this altered function may result in a clinical condition. Other variant polypeptides may find application in industry where a process may be improved or made more efficient by the presence of the variant. For example fermentation processes may be made more efficient following a mutation in a gene encoding a protein important for the process in question. Characterisation of that mutation may identify useful sites for additional or alternative mutations to further improve the process.
In addition there are numerous clinical examples of genetic mutation causing the expression of variant polypeptides with altered function. For example, many cancers have mutations in the p53 gene. Altered p53 function can dramatically affect a cell's ability to detect and eliminate genetic mutations, thus leaving an individual susceptible to cancer. There are many other examples, such as haemoglobinopathies where mutations within haemoglobin genes may result in clinical conditions such as α-thalassaemia. Sickle cell-anaemia, for example, results from a single point mutation in the gene encoding β-globin whereby the Glu-6(β) residue in Hb A is replaced by Val in sickle Hb (Hb S). It is thought that this hydrophobic side chain initiates a process by which the densely packed deoxyhaemoglobin tetramers inside the red cells interact with other side chains to form long polymeric fibres that distort the cells into a characteristic sickle shape. At least in theory if rapid analytical techniques could be developed these could be used in the diagnosis of disease states at an early stage allowing for early intervention strategies to be implemented.
Unfortunately many of the known analytical techniques used to analyse polypeptides are either not amenable to high throughput analysis or are such that they do not provide the required sensitivity to accurately distinguish between closely related polypeptides. As will be appreciated the ability to effectively distinguish between two closely related polypeptides is crucial. Without this ability any analytical technique is only capable of providing gross data on the polypeptides in the material studied. In addition many of the techniques are not sufficiently sensitive to be able to identify the presence of small amounts of polypeptide in very complex samples. This thus limits their usefulness.
Thus there remains a need for improved methods of analysing polypeptides to be developed, preferably ones which may be applicable in a clinical setting. Following significant research the present applicants identified MALDI-TOF mass spectrometry (MS) analysis as a diagnostic tool that showed promise. The present invention provides novel, rapid procedures utilising MALDI-TOF MS for analyzing polypeptides directly from a very small quantity of material. Thus, specific embodiments of the present invention provide methods useful for the clinical diagnosis of haemoglobinopathies as well as other diseases involving variant polypeptides.
In developing the improved methods the applicants also developed improved sample preparation techniques that were generally applicable to MALDI-TOF MS analysis of any material as well as being applicable to the improved methods and which provided improved outcomes. These improved sample preparation techniques typically provided improved sensitivity and sample to sample reproductity.
The discussion of documents, acts, materials, devices, articles and the like is included in this specification solely for the purpose of providing a context for the present invention. It is not suggested or represented that any or all of these matters formed part of the prior art base or were common general knowledge in the field relevant to the present invention as it existed in Australia before the priority date of each claim of this application.
As noted above the present invention relates to a number of improvements in relation to sample preparation techniques for MALDI-ToF MS analysis and the use of these sample preparation techniques in the analysis of polypeptides.
In a first aspect, the present invention provides a method of preparing a sample for MALDI-TOF MS analysis including the steps of:
The material to be analysed preferably includes a biological material or is derived from a biological material. Any biological material may be used including blood, cerebrospinal fluid, urine, saliva, seminal fluid or sweat or a combination thereof. It is preferred that the biological material is blood or derived from blood. Preferably the biological material includes a polypeptide. More preferably the polypeptide is a haemoglobin polypeptide or a fragment or variant or a haemoglobin peptide containing a covalently bonded adduct thereof. Preferably the haemoglobin polypeptide may include one or more of the following haemoglobins: α, β, γ, δ, ε or ζ. The biological material is obtained using techniques known in the art. The material may be applied to the carrier in any suitable form by techniques well known in the art. It is preferred that it is applied by a “spotting” technique. It is preferred that the biological material is diluted with a liquid preferably water prior to application. The liquid preferably contains a buffer such as ammonium bicarbonate buffer. The level of dilution will depend on the application but it is preferred that the dilution is from 1:10 to 1:10000. The amount of material applied is typically of the order of 0.1 to 1 0 μl, more preferably 0.5 to 5 μl, most preferably about 1 μl.
Following application of the material to be analysed at least a portion of the liquid component is removed. The liquid component may be removed in any suitable manner that does not destroy the integrity of compounds such as polypeptides within the material. For example the liquid may be removed by subjecting the applied material to elevated temperature, reduced pressure or a combination thereof. The liquid may also be removed by passing a stream of gas (preferably air) over the surface of the applied material. In a particularly preferred embodiment the liquid is removed by allowing the applied material to sit at ambient temperature and pressure for a sufficient time for the liquid to be removed by evaporation.
The amount of liquid removed may vary. It is preferred that at least 50% of the liquid component is removed, more preferably at least 75% of the liquid component is removed, yet even more preferably at least 90% of the liquid component is removed. In another preferred embodiment removal of the liquid component continues until the material is substantially dry, more preferably removal continues until the material is dry. Without wishing to be bound by theory it is felt that adequate removal of the liquid is important to minimise mixing between the material and the latter applied MALDI matrix layer. It is found that mixing of this type reduces the sensitivity of the later analysis.
Following the liquid removal step a MALDI matrix is applied using conventional techniques. Any suitable MALDI matrix may be used however it is preferred that the MALDI matrix is selected from the group consisting of sinapinic acid (SA), α-cyano-4-hydroxycinnamic acid (CHCA), 2,5-dihydroxybenzoic acid (2,5-DHB), 2-(4-hydroxy phenylazo)benzoic acid (HABA), succinic acid, 2,6-Dihydroxyacetophenone, Ferulic acid, caffeic acid, 2,4,6-trihydroxyacetophenone (THAP) and 3-hydroxypicolinic acid (HPA), Anthranilic acid, Nicotinic acid, Salicylamide and mixtures thereof. The amount of applied matrix may vary although it is typically of the order such that the ratio of matrix to material to be analysed is from 0.1:1 to 10:1, preferably from 0.5:1 to 5:1, most preferably 1:1 to 2:1.
The material to be analysed is preferably treated to partially digest polypeptides in the material. The digestion may be carried out in solution prior to application to a carrier or may be carried out after the material has been applied to the carrier. In one particularly preferred embodiment the material to be analysed is treated to partially digest polypeptides within the material prior to applying the material to the carrier. In this embodiment it is preferred that the digestion is carried out for from 1 to 24 hours, more preferably 4 to 24 hours. The treatment preferably includes contacting the material with a proteolytic agent. In another preferred embodiment the step of treating the material to partially digest polypeptides in the material is carried out on the carrier and preferably involves contacting the material to be analysed with a proteolytic agent. This may be achieved by addition of a proteolytic agent to the material after it has been applied to the carrier or by addition of a proteolytic agent to the carrier prior to addition of the material. The method preferably includes applying a proteolytic agent to the carrier prior to application of the material to be analysed such that following addition of the material the agent partially digests polypeptides within the material. In this embodiment the digestion is preferably carried out for a period of from 10 to 3600 seconds, more preferably 30 to 600 seconds, more preferably from 60 to 300 seconds, most preferably for 180 seconds.
Any suitable proteolytic agent may be used however it is preferred that the proteolytic agent is a protease, preferably a protease selected from the group consisting of trypsin and endoprotease Glu C. In one preferred embodiment the material is treated with a proteolytic agent in the presence of a surfactant. The surfactant is preferably sodium 3-[(2-methyl-2-undecyl-1,3-dioxolan-4-yl)methoxy]-1-propane-sulfonate.
The digestion is preferably allowed to continue until the digestion provides 100% sequence coverage of the polypeptide to be analysed. This can be readily determined by a skilled worker in the area. The digestion may be stopped in any way well known in the art. For example the digestion may be stopped by addition of a diluted acid. An example of a suitable acid is TFA.
In a second aspect, the present invention provides a method of preparing a sample for MALDI-ToF MS analysis, said sample including a material to be analysed and a carrier, the method including the step of conducting an on carrier digestion of polypeptides within the material.
The material to be analysed preferably includes a biological material or is derived from a biological material. Any biological material may be used in this aspect of the invention including blood, cerebrospinal fluid, urine, saliva, seminal fluid or sweat or a combination thereof. It is preferred that the biological material is blood. Preferably the biological material includes a polypeptide. More preferably the polypeptide is a haemoglobin polypeptide or a fragment or variant or a haemoglobin peptide containing a covalently bonded adduct thereof. Preferably the haemoglobin polypeptide may include one or more of the following haemoglobins: α, β, γ, δ, ε or ζ. The biological material is obtained using techniques well known in the art. The material may be applied to the carrier in any suitable form by techniques well known in the art. It is preferred that the material is applied by a spotting technique. It is preferred that the material is diluted with a liquid, preferably water, prior to applying it to the carrier. The liquid preferably contains a buffer such as ammonium bicarbonate. The level of dilution will depend on the application but it is preferred that the dilution is from 1:10 to 1:10000. The amount of material applied is typically of the order of 0.1 to 10 μl, more preferably 0.5 to 5.0 μl, most preferably about 1 μl. The method includes an on-carrier digest. The on-carrier digest preferably involves contacting the material with a proteolytic agent. This may be achieved by addition of a proteolytic agent to the carrier either prior to, simultaneously with, or following the addition of the material to be analysed.
The method preferably includes application of a proteolytic agent to the carrier prior to application of the material to be analysed such that following addition of the material to be analysed the agent partially digests polypeptides within the material. In this embodiment the digestion is preferably carried out for a period of from 10 to 3600 seconds, more preferably 30 to 600 seconds, more preferably from 60 to 300 seconds, most preferably for 180 seconds.
Any suitable proteolytic agent may be used however it is preferred that the proteolytic agent is a protease, preferably a protease selected from the group consisting of trypsin and endoprotease Glu C. In one preferred embodiment the material is treated with a proteolytic agent in the presence of a surfactant. The surfactant is preferably sodium 3-[(2-methyl-2-undecyl-1,3-dioxolan-4-yl)methoxy]-1-propane-sulfonate.
The digestion is preferably allowed to continue until the digestion provides 100% sequence coverage of the polypeptide to be analysed. The digestion may be stopped in any way well known in the art. For example the digestion may be stopped by addition of a diluted acid. An example of a suitable acid is TFA.
A particularly preferred way of terminating the digestion is by applying a MALDI matrix over the material. Any suitable MALDI matrix may be used however the MALDI matrix is preferably selected from the group consisting of sinapinic acid (SA), α-cyano-4-hydroxycinnamic acid (CHCA), 2,5-dihydroxybenzoic acid (2,5-DHB), 2-(4-hydroxyphenylazo)benzoic acid (HABA), succinic acid, 2,6-Dihydroxyacetophenone, Ferulic acid, caffeic acid, 2,4,6-trihydroxy acetophenone (THAP) and 3-hydroxypicolinic acid (HPA), Anthranilic acid, Nicotinic acid, Salicylamide or mixtures thereof. The amount of applied matrix may vary although it is typically of the order such that the ratio of matrix to sample is from 0.1:1 to 10:1, preferably 0.5:1 to 5:1, most preferably 1:1 to 2:1.
In a third aspect, the present invention provides a sample for analysis having,
(a) a carrier having a surface;
(b) a layer including a material to be analysed, and
(c) a single MALDI matrix layer,
wherein the layer including the material to be analysed is located between the carrier surface and the MALDI matrix layer.
The material to be analysed preferably includes a biological material or is derived from a biological material. Any biological materials may be used including blood, cerebrospinal fluid, urine, saliva, seminal fluid or sweat or a combination thereof. It is preferred that the biological material is blood. Preferably the biological material includes a polypeptide. More preferably the polypeptide is a haemoglobin polypeptide or a fragment or variant or a haemoglobin peptide containing a covalently bonded adduct thereof. Preferably the haemoglobin polypeptide may include one or more of the following haemoglobins: α, β, γ, δ, ε or ζ. It is particularly preferred that the material to be analysed contains partially digested polypeptides.
Any suitable-MALDI matrix may be utilised however it is preferred that the MALDI matrix is selected from the group consisting of sinapinic acid (SA), α-cyano-4-hydroxycinnamic acid (CHCA), 2,5-dihydroxybenzoic acid (2,5-DHB), 2-(4-hydroxyphenylazo)benzoic acid (HABA), succinic acid, 2,6-Dihydroxyacetophenone, Ferulic acid, caffeic acid, 2,4,6-trihydroxyacetophenone (THAP) and 3-hydroxypicolinic acid (HPA), Anthranilic acid, Nicotinic acid, Salicylamide and mixtures thereof. It is preferred that the sample has been produced using the methods of the invention described herein.
In a fourth aspect, the present invention provides a method of improving digestion of polypeptides within a material said method including the step of conducting the digestion in the presence of sodium 3-[(2-methyl-2-undecyl-1,3-dioxolan-4-yl)methoxy]-1-propane-sulfonate or a derivative thereof. In a preferred embodiment the digestion includes digestion by proteolytic enzymes.
In a fifth aspect the invention provides a method of analysing a polypeptide including the steps of:
The step of partially digesting the polypeptide is preferably carried out by contacting the polypeptide with a proteolytic agent. The digestion may be carried out in solution prior to application to a carrier or may be carried out after the material has been applied to the carrier. Accordingly, the polypeptide may be digested either in solution or whilst on a carrier. In one preferred embodiment the digestion is carried out in solution by addition of a proteolytic agent to a solution containing the polypeptide. In this embodiment it is preferred that the digestion is carried out for from 1 to 24 hours, preferably from 4 to 24 hours. Following digestion the material is typically applied to the carrier. The amount of material applied is typically of the order of 0.1 to 10 μl, more preferably 0.5 to 5 μl, most preferably about 1 μl.
Following application of the material to be analysed at least a portion of the liquid component is removed. The liquid component may be removed in any suitable manner that does not destroy the integrity of compounds such as polypeptides within the material. For example the liquid may be removed by subjecting the applied material to elevated temperature, reduced pressure or a combination thereof. The liquid may also be removed by passing a stream of gas (preferably air) over the surface of the applied material. In a particularly preferred embodiment the liquid is removed by allowing the applied material to sit at ambient temperature and pressure for a sufficient time for the liquid to be removed by evaporation.
The amount of liquid removed may vary. It is preferred that at least 50% of the liquid component is removed, more preferably at least 75% of the liquid component is removed, yet even more preferably at least 90% of the liquid component is removed. In another preferred embodiment removal of the liquid component continues until the material is substantially dry, more preferably removal continues until the material is dry. Without wishing to be bound by theory it is felt that adequate removal of the liquid is important to minimise mixing between the material and the latter applied MALDI matrix layer. It is found that mixing of this type reduces the sensitivity of the later analysis.
Following the liquid removal step a MALDI matrix is applied using conventional techniques. Any suitable MALDI matrix may be used however it is preferred that the MALDI matrix is selected from the group consisting of sinapinic acid (SA), α-cyano-4-hydroxycinnamic acid (CHCA), 2,5-dihydroxybenzoic acid (2,5-DHB), 2-(4-hydroxy phenylazo)benzoic acid (HABA), succinic acid, 2,6-Dihydroxyacetophenone, Ferulic acid, caffeic acid, 2,4,6-trihydroxyacetophenone (THAP) and 3-hydroxypicolinic acid (HPA), Anthranilic acid, Nicotinic acid, Salicylamide and mixtures thereof. The amount of applied matrix may vary although it is typically of the order such that the ratio of matrix to material to be analysed is from 0.1:1 to 10:1, preferably from 0.5:1 to 5:1, most preferably 1:1 to 2:1.
In another preferred embodiment the digestion is carried out on a carrier. In this embodiment the method preferably includes applying a proteolytic agent to a carrier prior to application of the polypeptide to the carrier such that following addition of the material the agent partially digests the polypeptide. In this embodiment the digestion is preferably carried out for a period of from 10 to 3600 seconds, more preferably 30 to 600 seconds, more preferably from 60 to 300 seconds, most preferably for 180 seconds.
Any suitable proteolytic agent may be used however it is preferred that the proteolytic agent is a protease, preferably a protease selected from the group consisting of trypsin and endoprotease Glu C. It is preferred that the material is treated with a proteolytic agent in the presence of a surfactant. The surfactant is preferably sodium 3-[(2-methyl-2-undecyl-1,3-dioxolan-4-yl)methoxy]-1-propane-sulfonate.
The digestion is preferably allowed to continue until the digestion provides 100% sequence coverage of the polypeptide to be analysed. The digestion may be stopped in any way well known in the art. For example the digestion may be stopped by addition of an acid. An example of a suitable acid is TFA. A particularly preferred way of terminating the digestion of the on carrier digest is by applying a MALDI matrix over the material. Any suitable MALDI matrix may be used however the MALDI matrix is preferably selected from the group consisting of sinapinic acid (SA), α-cyano-4-hydroxycinnamic acid (CHCA), 2,5-dihydroxybenzoic acid (2,5-DHB), 2-(4-hydroxyphenylazo) benzoic acid (HABA), succinic acid, 2,6-Dihydroxyacetophenone, Ferulic acid, caffeic acid, 2,4,6-trihydroxy acetophenone (THAP) and 3-hydroxypicolinic acid (HPA), Anthranilic acid, Nicotinic acid, Salicylamide or mixtures thereof.
The analysis of the MALDI-ToF MS output is conducted in any way well known in the art. It is preferred, however, that the analysis is such that a sequence window is chosen to ensure that fragments exist which cover the entire sequence of the polypeptide. Analysis of this window can then be used to determine digestion fragments characteristic of the polypeptide. Fragments of this type are effectively “signature” fragments and may be indicative of the presence of the polypeptide in a complex mixture that has been digested in a similar manner. The data obtained from such analysis can be added to a database or library of fragments for use in the later identification of the presence of the polypeptide in complex mixtures.
In yet an even further aspect the invention provides a method of determining the identity of one or more polypeptide(s) in a material including the steps of:
The material preferably includes a biological material or is derived from a biological material. A number of biological materials may be used including blood, cerebrospinal fluid, urine, saliva, seminal fluid or sweat or a combination thereof. It is preferred that the biological material is blood. Preferably the biological material includes a polypeptide. More preferably the polypeptide is a haemoglobin polypeptide or a fragment or variant or a haemoglobin peptide containing a covalently bonded adduct thereof. Preferably the haemoglobin polypeptide may include one or more of the following haemoglobins: α, β, γ, δ, ε or ζ. It is particularly preferred that the material to be analysed contains partially digested polypeptides.
The step of partially digesting the material preferably involves contacting the material with a proteolytic agent. The digestion may be carried out in solution prior to application to a carrier or may be carried out after the material has been applied to the carrier. Accordingly, the material may be digested either in solution or whilst on a carrier. In one preferred embodiment the digestion is carried out in solution by addition of a proteolytic agent to a solution containing the material. In this embodiment it is preferred that the digestion is carried out for from 1 to 24 hours, more preferably 4 to 24 hours. Any suitable proteolytic agent may be used in the digestion however it is preferred that the proteolytic agent is a protease, preferably a protease selected from the group consisting of trypsin and endoprotease Glu C. In one preferred embodiment the digestion is conducted in the presence of a surfactant. The surfactant is preferably sodium 3-[(2-methyl-2-undecyl-1,3-dioxolan-4-yl)methoxy]-1-propane-sulfonate. The digestion may be stopped by any method well known in the art. Following the in solution digestion the digested material is preferably applied to a carrier.
In this embodiment following application of the material at least a portion of the liquid component is removed. The liquid component may be removed in any suitable manner that does not destroy the integrity of polypeptides or polypeptide fragments within the material. For example the liquid may be removed by subjecting the applied material to elevated temperature, reduced pressure or a combination thereof. The liquid may also be removed by passing a stream of gas (preferably air) over the surface of the applied material. In a particularly preferred embodiment the liquid is removed by allowing the applied material to sit at ambient temperature and pressure for a sufficient time for the liquid to be removed by evaporation.
The amount of liquid removed may vary. It is preferred that at least 50% of the liquid component is removed, more preferably at least 75% of the liquid component is removed, yet even more preferably at least 90% of the liquid component is removed. In another preferred embodiment removal of the liquid component continues until the material is substantially dry, more preferably removal continues until the material is dry. Following the liquid removal step a MALDI matrix is applied. Any suitable MALDI matrix may be used however it is preferred that the MALDI matrix is selected from the group consisting of sinapinic acid (SA), α-cyano-4-hydroxycinnamic acid (CHCA), 2,5-dihydroxybenzoic acid (2,5-DHB), 2-(4-hydroxyphenylazo)benzoic acid (HABA), succinic acid, 2,6-Dihydroxyacetophenone, Ferulic acid, caffeic acid, 2,4,6-trihydroxy acetophenone (THAP) and 3-hydroxypicolinic acid (HPA), Anthranilic acid, Nicotinic acid, Salicylamide and mixtures thereof. The amount of MALDI matrix may vary being typically of the order such that the ratio of matrix to added sample is from 0.1 to 1 to 10:1, preferably 0.5:1 to 5:1, most preferably from 1:1 to 2:1.
In another preferred embodiment the digestion is carried out on a carrier. This may be carried out by applying a proteolytic agent either prior to, simultaneously with, or after the application of the material to be analysed. In this embodiment the method preferably includes applying a proteolytic agent to a carrier prior to application of the material to the carrier such that following addition of the material the agent partially digests any polypeptides within the material. In this embodiment the digestion is preferably carried out for a period of from 10 to 3600 seconds, more preferably 30 to 600 seconds, more preferably from 60 to 300 seconds, most preferably for 180 seconds.
Any suitable proteolytic agent may be used in the digestion however it is preferred that the proteolytic agent is a protease, preferably a protease selected from the group consisting of trypsin and endoprotease Glu C. In one preferred embodiment digestion occurs in the presence of a surfactant. The surfactant is preferably sodium 3-[(2-methyl-2-undecyl-1,3-dioxolan-4-yl)methoxy]-1-propane-sulfonate.
The digestion is preferably allowed to continue until the digestion provides 100% sequence coverage of the polypeptide to be analysed for. The digestion may be stopped in any way well known in the art. For example the digestion may be stopped by addition of a diluted acid either to the digestion in solution or to the on carrier digestion. An example of a suitable acid is TFA. A particularly preferred way of terminating the on carrier digestion is by applying a MALDI matrix over the material. Any suitable MALDI matrix may be used however the MALDI matrix is preferably selected from the group consisting of sinapinic acid (SA), α-cyano-4-hydroxycinnamic acid (CHCA), 2,5-dihydroxybenzoic acid (2,5-DHB), 2-(4-hydroxyphenylazo)benzoic acid (HABA), succinic acid, 2,6-Dihydroxyacetophenone, Ferulic acid, caffeic acid, 2,4,6-trihydroxyacetophenone (THAP) and 3-hydroxypicolinic acid (HPA), Anthranilic acid, Nicotinic acid, Salicylamide or mixtures thereof.
Following the production of the sample by the methods described above it is then subjected to analysis by MALDI-TOF MS to determine digestion fragments for the material. The digestion fragments are typically indicative of the polypeptides in the original material. Once the digestion fragments have been determined they are compared to the known digestion fragments (typically called the signature fragments) of known polypeptides. The comparison of the digestion fragments with known digestion fragments or with “signature” digestion fragments of known polypeptides may be carried out in any of a number of ways. For example this can be done manually by scanning the output of the MALDI-TOF MS and comparing it to known digestion fragments to determine the identity of one or more of the polypeptides present. It is preferred that the comparison is carried out by computerised means. In a particularly preferred embodiment the output of the MALDI-TOF MS analysis is compared by computer means to a library of signature fragments to identify a plurality of polypeptides in the material.
In a particularly preferred embodiment the method is used to determine the presence of a polypeptide in a sample. In this embodiment the digestion fragments are compared with the “signature” digestion fragments of the polypeptide of interest to determine if that particular polypeptide is present. This method therefore allows for the determination of the presence of a polypeptide of interest in a complex mixture of polypeptides.
In yet an even further aspect the invention provides a method of analysing a polypeptide variant including the steps of:
The material preferably includes a biological material or is derived from a biological material. Any biological materials may be used including blood, cerebrospinal fluid, urine, saliva, seminal fluid or sweat or a combination thereof. It is preferred that the biological material is blood. Preferably the biological material includes a polypeptide. More preferably the polypeptide is a haemoglobin polypeptide or a fragment or variant or a haemoglobin peptide containing a covalently bonded adduct thereof. Preferably the haemoglobin polypeptide may include one or more of the following haemoglobins: α, β, γ, δ, ε or ζ. It is particularly preferred that the material to be analysed contains partially digested polypeptides.
The digestion preferably involves contacting the material with a proteolytic agent. The digestion may be carried out in solution prior to application to a carrier or may be carried out after the material has been applied to the carrier. Accordingly, the material may be digested either in solution prior to application to the carrier or whilst on a carrier. In one preferred embodiment the digestion is carried out in solution by addition of a proteolytic agent to a solution containing the material. In this embodiment it is preferred that the digestion is carried out for from 1 to 24 hours, more preferably from 4 to 24 hours. Any suitable proteolytic agent may be used in the digestion however it is preferred that the proteolytic agent is a protease, preferably a protease selected from the group consisting of trypsin and endoprotease Glu C. In one preferred embodiment the digestion is carried out in the presence of a surfactant. The surfactant is preferably sodium 3-[(2-methyl-2-undecyl-1,3-dioxolan-4-yl)methoxy]-1-propane-sulfonate. The digestion may be stopped by any method well known in the art. In this embodiment following the in solution digestion the digested material is preferably added to a carrier.
Following application of the material to the carrier at least a portion of the liquid component is removed. The liquid component may be removed in any suitable manner that does not destroy the integrity of polypeptides or polypeptide fragments within the material. For example the liquid may be removed by subjecting the applied material to elevated temperature, reduced pressure or a combination thereof. The liquid may also be removed by passing a stream of gas (preferably air) over the surface of the applied material. In a particularly preferred embodiment the liquid is removed by allowing the applied material to sit at ambient temperature and pressure for a sufficient time for the liquid to be removed by evaporation.
The amount of liquid removed may vary. It is preferred that at least 50% of the liquid component is removed, more preferably at least 75% of the liquid component is removed, yet even more preferably at least 90% of the liquid component is removed. In another preferred embodiment removal of the liquid component continues until the material is substantially dry, more preferably removal continues until the material is dry. Following the liquid removal step a MALDI matrix is applied. Any suitable MALDI matrix may be used however it is preferred that the MALDI matrix is selected from the group consisting of sinapinic acid (SA), α-cyano-4-hydroxycinnamic acid (CHCA), 2,5-dihydroxybenzoic acid (2,5-DHB), 2-(4-hydroxyphenylazo)benzoic acid (HABA), succinic acid, 2,6-Dihydroxyacetophenone, Ferulic acid, caffeic acid, 2,4,6-trihydroxy acetophenone (THAP) and 3-hydroxypicolinic acid (HPA), Anthranilic acid, Nicotinic acid, Salicylamide and mixtures thereof. The amount of applied matrix may vary although it is typically of the order such that the ratio of matrix to added sample is from 0.1:1 to 10:1, preferably from 0.5:1 to 5:1, most preferably from 1:1 to 2:1.
In another preferred embodiment the digestion is carried out on a carrier. In this embodiment the method preferably includes applying a proteolytic agent to a carrier prior to application of the material to the carrier such that following addition of the material the agent partially digests any polypeptides within the material. In this embodiment the digestion is preferably carried out for a period of from 10 to 3600 seconds, more preferably 30 to 600 seconds, more preferably from 60 to 300 seconds, most preferably for 180 seconds.
Any suitable proteolytic agent may be used in the digestion however it is preferred that the proteolytic agent is a protease, preferably a protease selected from the group consisting of trypsin and endoprotease Glu C. In one preferred embodiment the digestion occurs in the presence of a surfactant. The surfactant is preferably sodium 3-[(2-methyl-2-undecyl-1,3-dioxolan-4-yl)methoxy]-1-propane-sulfonate.
The digestion is preferably allowed to continue until the digestion provides 100% sequence coverage of the polypeptide to be analysed for. The digestion may be stopped in any way well known in the art. For example the digestion may be stopped by addition of a diluted acid either to the digestion in solution or to the on carrier digestion. An example of a suitable acid is TFA. A particularly preferred way of terminating the on carrier digestion is by applying a MALDI matrix over the material. Any suitable MALDI matrix may be used however the MALDI matrix is preferably selected from the group consisting of sinapinic acid (SA), α-cyano-4-hydroxycinnamic acid (CHCA), 2,5-dihydroxybenzoic acid (2,5-DHB), 2-(4-hydroxyphenylazo)benzoic acid (HABA), succinic acid, 2,6-Dihydroxyacetophenone, Ferulic acid, caffeic acid, 2,4,6-trihydroxyacetophenone (THAP) and 3-hydroxypicolinic acid (HPA), Anthranilic acid, Nicotinic acid, Salicylamide or mixtures thereof.
The digested material is subjected to analysis by MALDI-ToF MS to determine digestion fragments for the material. Once the digestion fragments have been determined they are compared to the known digestion fragments (typically called the signature fragments) of the non variant polypeptides. Whilst this can be done manually by scanning the output of the MALDI-TOF MS and comparing it to digestion fragments of known non-variant polypeptides it is preferred that the comparison is carried out by computerised means. In a particularly preferred embodiment the output of the MALDI-ToF MS analysis is compared by computer means to a library of signature fragments for non variant polypeptides to determine the fragment containing the variation. Once the fragment has been determined it is generally straightforward to determine the nature of the variation.
In yet a further aspect the invention provides a method of diagnosing a condition in a subject including the steps of:
The condition to be diagnosed is either a condition that is characterised by the absence of a polypeptide that would be present in material obtained from a non-afflicted subject or a condition that is characterised by the presence in the material of a polypeptide characteristic of the condition, said polypeptide not being present in a sample of a non-afflicted subject. In a preferred embodiment the condition is a haemoglobinopathy. Haemoglobinopathies fall into overlapping groups: thalassemias (imbalance in globinchain production) and haemoglobin variants (structurally abnormal haemoglobins). Haemoglobinopathoies include: alpha-thalassemia (non-deletional, deletional, Hb H disease), beta-thalassemia, delta-thalassemia, gamma-thalassemia, hereditary persistence of fetal hemoglobin (HPFH), deltabeta-thalassemia, sickle cell disorder and other haemoglobin variant related disorders.
In principle the material obtained may be any bodily material or extract. Examples of materials that may be used include blood, CSF fluid, urine, saliva, seminal fluid or sweat or a combination thereof. It is preferred that the material is blood. The material is obtained from the subject using standard techniques well known in the art.
The material is then analysed by MALDI-ToF MS to determine polypeptides in the material. The analysing step preferably involves subjecting the material to be analysed to MALDI-ToF MS analysis on a carrier. The material on the carrier has preferably been subjected to a partial digestion.
The digestion may be carried out in solution prior to application to a carrier or may be carried out after the material has been applied to the carrier. Accordingly, the material may be digested either in solution or whilst on the carrier. In one preferred embodiment the digestion is carried out in solution by addition of a proteolytic agent to a solution containing the material. In this embodiment it is preferred that the digestion is carried out for from 1 to 24 hours, more preferably 4 to 24 hours. Any suitable proteolytic agent may be used in the digestion however it is preferred that the proteolytic agent is a protease, preferably a protease selected from the group consisting of trypsin and endoprotease Glu C. In one preferred embodiment the material is digested in the presence of a surfactant. The surfactant is preferably sodium 3-[(2-methyl-2-undecyl-1,3-dioxolan-4-yl)methoxy]-1-propane-sulfonate. The digestion may be stopped by any method well known in the art. Following the in solution digestion the digested material is then preferably applied to a carrier.
Following application of the material to the carrier at least a portion of the liquid component is removed. The liquid component may be removed in any suitable manner that does not destroy the integrity of polypeptides within the material. For example the liquid may be removed by subjecting the applied material to elevated temperature, reduced pressure or a combination thereof. The liquid may also be removed by passing a stream of gas (preferably air) over the surface o the applied material. In a particularly preferred embodiment the liquid is removed by allowing the applied material to sit at ambient temperature and pressure for a sufficient time for the liquid to be removed by evaporation.
The amount of liquid removed may vary. It is preferred that at least 50% of the liquid component is removed, more preferably at least 75% of the liquid component is removed, yet even more preferably at least 90% of the liquid component is removed. In another preferred embodiment removal of the liquid component continues until the material is substantially dry, more preferably removal continues until the material is dry. Following the liquid removal step a MALDI matrix is applied. Any suitable MALDI matrix may be used however it is preferred that the MALDI matrix is selected from the group consisting of sinapinic acid (SA), α-cyano-4-hydroxycinnamic acid (CHCA), 2,5-dihydroxybenzoic acid (2,5-DHB), 2-(4-hydroxyphenylazo)benzoic acid (HABA), succinic acid, 2,6-Dihydroxyacetophenone, Ferulic acid, caffeic acid, 2,4,6-trihydroxy acetophenone (THAP) and 3-hydroxypicolinic acid (HPA), Anthranilic acid, Nicotinic acid, Salicylamide and mixtures thereof.
In another preferred embodiment the digestion is carried out on the carrier. In this embodiment the method preferably includes applying a proteolytic agent to a carrier prior to application of the material to the carrier such that following addition of the material the agent partially digests any polypeptides within the material. In this embodiment the digestion is preferably carried out for a period of from 10 to 3600 seconds, more preferably 30 to 600 seconds, more preferably from 60 to 300 seconds, most preferably for 180 seconds.
Any suitable proteolytic agent may be used in the digestion however it is preferred that the proteolytic agent is a protease, preferably a protease selected from the group consisting of trypsin and endoprotease Glu C. In one preferred embodiment the digestion in the presence of a surfactant. The surfactant is preferably sodium 3-[(2-methyl-2-undecyl-1,3-dioxolan-4-yl)methoxy]-1-propane-sulfonate.
The digestion is preferably allowed to continue until the digestion provides 100% sequence coverage of the polypeptide to be analysed for. The digestion may be stopped in any way well known in the art. For example the digestion may be stopped by addition of a diluted acid either to the digestion in solution or to the on carrier digestion. An example of a suitable acid is TFA. A particularly preferred way of terminating the on carrier digestion is by applying a MALDI matrix over the material. Any suitable MALDI matrix may be used however the MALDI matrix is preferably selected from the group consisting of sinapinic acid (SA), α-cyano-4-hydroxycinnamic acid (CHCA), 2,5-dihydroxybenzoic acid (2,5-DHB), 2-(4-hydroxyphenylazo)benzoic acid (HABA), succinic acid, 2,6-Dihydroxyacetophenone, Ferulic acid, caffeic acid, 2,4,6-trihydroxacetophenone (THAP) and 3-hydroxypicolinic acid (HPA), Anthranilic acid, Nicotinic acid, Salicylamide or mixtures thereof.
Once the sample has been prepared in the manner discussed above it is subjected to MALDI ToF MS analysis using standard operating conditions. The MALDI-TOF MS output is then analysed to determine from the digestion fragments the identity of one or more polypeptides within the material. The diagnosis of the condition is then based on the presence or absence of a polypeptide from the material. The output may be analysed using any of a number of techniques. At its most simplistic the output may be viewed manually to determine the digestion fragments and to determine if signature digestion fragments are present. It is preferred, however, that the output is compared using computer aided techniques with a database or library of known fragments. Any significant mass/charge signal representing a peptide, which is different from haemoglobin A, may constitute a Haemoglobin variant. If this variant is associated with a clinical significant characteristic it constitutes a haemoglobinopathy.
The term “polypeptide” refers to a chain of amino acids, wherein adjacent amino acids are linked by peptide bonds. The amino acids may be naturally occurring amino acids or modified amino acids. Other terms such as “protein” or “peptide” are intended to be encompassed by the term “polypeptide”.
The methods of sample preparation and analysis of the present invention are applicable to a wide range of materials, however it is preferred that the materials include biological materials or are derived from biological materials. In a particularly preferred embodiment the material is a biological material.
Any suitable biological material may be used, however it is preferred that the biological material is selected from the group consisting of blood, cerebrospinal fluid, urine, saliva, seminal fluid, sweat and a combination thereof. These samples may be obtained using techniques well known in the art that need no further elaboration.
Once obtained the material is then typically diluted in a liquid, preferably water. The liquid preferably includes a buffer. A suitable buffer is ammonium bicarbonate and a suitable level of dilution is from 1:10 to 1:10000. This is found to provide a suitable level of material for further analysis by the techniques described herein.
As stated previously the sample preparation techniques and methods of analysis as described herein provide improvements in the performance of the analysis of the sample. They typically provide improved sensitivity and/or reproducibility of the analysis.
The sample preparation techniques and methods of analysis as described herein typically involve addition of a material to a carrier. The amount of material added may vary considerably depending on the final application but it is typically of the order of 0.1 to 10 μl, more preferably 0.5 to 5.0 μl, most preferably about 1 μl. Any carrier well known in the art may be used. Examples of suitable carriers include Stainless steel carrier plates, gold carrier plates, carrier plates with hydrophobic surfaces, carrier plates with surface indentations (used with gel membranes).
In a particularly preferred embodiment the carriers have a plurality of sample positions such that a plurality of samples may be added to the one carrier. This allows for rapid throughput analysis of a number of samples on a MALDI-ToF MS apparatus and therefore provides for an economic process to be carried out.
In order to perform the methods of the invention as described herein it is preferable to digest the material to be analysed so that any polypeptides in the material are cleaved into smaller peptides which are more amenable to MALDI-ToF MS analysis. For the methods of the present invention, the applicants have found that a partial digest is able to give rapid and consistently accurate analysis of the material to be analysed.
The optimal conditions under which the partial digest is carried out must be determined for each class of polypeptides to be analysed and will depend on the material to be analysed. The skilled addressee will readily understand how to perform test digests in order to determine suitable conditions. Details of such digests are described below in reference to haemoglobins and are illustrative of the method to be used on a use by use basis. Conditions that need to be considered include, but are not limited to, enzyme, buffer, temperature, polypeptide concentration and time of digestion.
The digestion may be carried out either in solution or on a carrier, or a combination thereof. The digestion typically involves contacting the material with a proteolytic material. There are a large number of proteolytic materials well known in the art and the appropriate proteolytic material will depend upon the polypeptides expected to be present in the material to be digested. In general a skilled worker will be able to select a suitable proteolytic material with little difficulty. The amount of proteolytic material to be used will depend on the speed of digestion required. Once again through routine experimentation this can be readily determined.
The digestion may be carried out prior to application to a carrier. In this embodiment the digestion is typically carried out in solution. The digestion typically is carried out for a period of time suitable to provide at least a partial digestion of the polypeptides. The length of time will vary based on the polypeptides present but is typically from 4 to 24 hours. The digestion is typically carried out at temperatures well known in the art, generally from 0 to 100° C., more preferably 10 to 75° C. The exact temperature chosen will depend on the nature of the proteolytic agent and its optimal temperature range.
The digestion may be stopped using any technique well known in the art. Exemplary of such a technique is the addition of an acid. The material is then applied to a carrier as described previously herein.
The material is preferably applied by a spotting technique which would be well known to a skilled worker in the field.
After the material has been applied to the carrier it is typical that a MALDI matrix is applied using standard techniques. Any suitable MALDI matrix may be used but it is preferably selected from the groups described previously.
The sample is then analysed using standard MALDI-TOF techniques to determine the digestive fragments of the material to be analysed.
In particular embodiments of the present invention the partial digestion may be performed in the presence of a surfactant. Preferably, the surfactant is sodium 3-[(2-methyl-2-undecyl-1,3-dioxolan-4-yl)methoxy]-1-propane-sulfonate.
The methods of the invention all involve analysis on the basis of characteristic fragments of the polypeptide of interest. These characteristic fragments are commonly known as “signature” fragments of the polypeptide. There are a number of advantages in analysing a partially digested material for the presence of signature fragments of this type. The principal advantage is that in general the presence or absence of a signature fragment is determinative of whether the polypeptide is present or absent. This is generally more reliable than analysing the undigested material as the resolution with non-fragmented samples is not as great. Accordingly the use of fragmentation analysis therefore provides significant advantages.
In general for a large number of polypeptides the signature digestion peptides are known from the art and libraries of such peptides are available. In circumstances where the signature peptides are not known it is relatively straightforward to determine their identity. This can either be done theoretically based on the expected cleavage points of the polypeptide (which will be determined by the proteolytic agent of interest) or by subjecting a standard sample of the polypeptide to digestion conditions followed by analysis to determine the signature peptides. In general therefore the signature peptides can be determined quite easily either by theoretical means or by routine experimentation. If experimentation is used it is preferable to use the same conditions in the determination of the signature fragments as will be used in the material analysis.
Of course, once the signature fragments of a number of polypeptides are known this information can be used in methods of determining the identity of polypeptides in a material. Accordingly, if a material is subjected to digestion and then analysed the output of the MALDI-TOF mass spectrometry will provide the digestion fragments of the polypeptides in the material. Comparison of this output to the signature peptides of the known polypeptides (preferably by computer) allows for the identification of many of the polypeptides in the material. This allows for the rapid analysis of a complex material containing a number of polypeptides.
A particularly preferred use of this methodology is to determine if a material contains a particular polypeptide of interest. This can be very useful as the presence of the polypeptide may be indicative of a medical condition. This involves comparison of the MALDI-TOF MS output with the signature peptide or peptides of the polypeptide of interest. If the signature peptide is present this is indicative of the presence of the polypeptide of interest.
The fragment analysis discussed above can also be used in polypeptide variant analysis. By comparing the fragmentation pattern of a polypeptide variant with the fragmentation pattern of non-variant polypeptides it is generally easy to determine the fragment containing the variation (as it will be new). Once this has been done analysis of the difference between the new fragment and the corresponding non-variant fragment can be used to determine the difference in the variant.
The analysis of polypeptide variants in this way of course provides the analyst with signature peptides of the polypeptide variant which can be used as further probes for the presence of that polypeptide variant in complex mixtures Finally, the ability to accurately analyse complex materials for the presence or absence of a polypeptide may be a useful diagnostic tool.
A number of medical conditions are characterised by a gene defect such that the gene is not expressed in the body. The direct physiological effect of this non-expression of the gene is the absence in the body of the polypeptides that would be expressed in the body of a person without the gene defect. Accordingly the ability to accurately analyse a biological sample for the absence of a polypeptide may be used diagnostically. This is done by analysing the output and determining if the signature fragment of the polypeptide is present. If the signature fragment is not present it can be concluded that the polypeptide was not present in the sample further indicating that the individual had the gene defect. Alternatively, quantitative data can be used to determine if the polypeptide is present but at a reduced level (in some instances the gene defect leads to reduced production of the polypeptide).
In a number of other conditions there is not the absence of gene expression, rather the gene produces a polypeptide variant that is indicative of the condition. In these instances it is more reliable to analyse the individual for the presence of the polypeptide variant which will not be present in a sample from a healthy patient. This is because in some clinical conditions the person produces a certain amount of the “normal” polypeptide as well as an amount of the polypeptide variant. Merely analysing the sample for the absence of the normal polypeptide would therefore not be conclusive.
The method may be applied to any condition (typically a genetic defect) which is manifested in the production of an abnormal polypeptide (or a polypeptide variant). In many instances the presence of variant polypeptides is well known in the art and the present invention provides an improved method for the rapid qualitative analysis of these variants. Once the presence of the variant has been confirmed (by the presence of the signature fragment of the variant) the diagnosis of the condition that the presence of that variant indicates can be made.
One family of conditions that can be diagnosed using this technique are haemoglobenopathies which are manifested in variations in the α and β globin chains. In this family in general the known haemoglobenopathies are well documented and the polypeptides characteristic of each haemoglobenopathy well characterised. As such analysis for the presence of the polypeptides can be used in the diagnosis of the particular haemoglobenopathy.
In order to demonstrate the applicability of the improved sample preparation techniques and analytical methods, haemoglobins have been analysed as an indicative class of polypeptides. While the Examples below concentrate on haemoglobins, the skilled addressee would readily understand the methodology explained and be able to apply the methods to other polypeptide systems. Thus the choice of haemoglobins is intended demonstrate the applicability of the methodology and in no way is intended to limit the scope of the present invention.
Haemoglobinopathies are a major public health problem causing significant ill health, disability and death among the world populations. It has been estimated that at least 7% of the world's population are carriers of haemoglobinopathies. With the completion of the human genome project attention has now turned to studies of genetic diseases and their contribution to ill health and suffering in the community. In multicultural societies such as Australia screening for haemoglobinopathies is of increasing public health importance. Methods for diagnosis and management of these conditions need to be simpler, more rapid and more cost effective.
In general the polypeptide analysis techniques that are currently available are typically slow and not suited to fast throughput analysis. This can be seen by reference to the diagnostic approaches employed to detect haemoglobinopathies. The utility of the different methods currently used depends on the intended purpose, the availability of resources and the type of available technology. Initial and follow-up tests in practice include full blood examination (FBE), solubility and sickling tests, HbA2 and HbF quantification and determination of the ferritin level, currently being performed by electrophoresis, iso-electric focusing (IEF), high-performance liquid chromatography (HPLC) and DNA analysis. Detection of ζ-globin chains in the cord blood by enzyme-linked immunoassay (ELISA) for screening for α-thalassaemia has also been described.
Many of the heterozygous and homozygous states for haemoglobin (Hb) disorders do not change the red cell morphology. Clinically significant Hb variants are usually first observed by routine haematological procedures. A low Hb level, microcytosis, hypochromia, blood film findings (target cells, fragmented red blood cells (RBCs), nucleated RBCs) are useful for the detection of thalassaemia major, sickle cell disease and unstable Hbs and are still the main screening tool in many of the poorer third world countries. Red cell indices are used to screen for β- and α-thalassaemia carrier states. Low mean corpuscular volume (MCV) (<82 fL) and mean corpuscular haemoglobin (MCH) (<27 pg) are indicative of such cases when iron deficiency is excluded even though the blood Hb level may not be lower than normal. Haemolysis is indicated by raised reticulocyte count. Reticulocyte count is also useful to provide information on unstable Hb variants, HbH disease or sickle cell disease. A high Hb level and increased haematocrit (HCT) level indicate erythrocytosis, which along with appropriate clinical observations may suggest a Hb variant with high oxygen affinity. Although these methods have their merits in the clinical diagnostics, they provide mainly morphological descriptions, which give extremely limited information on Hb variants.
While the cell observation techniques described above can assist in the identification of the presence of a haemoglobinopathy, they cannot identify the particular haemoglobin variant present. Molecular studies are required to identify the haemoglobin variant, which in turn may allow specific treatment of a patient.
Electrophoresis is one of the oldest methods available for the screening for Hb variants, and typically is used to screen a small number of samples. It has been used for detection and quantification of Hb variants. Because different haemoglobins may migrate similarly under a given set of conditions, electrophoresis is usually performed at two different pH values and on two different supporting mediums. The usual choice is cellulose acetate electrophoresis at pH 8.4 and citrate agar electrophoresis at pH 6.0. Most laboratories use commercially available kits that allow both medium and pH (6.0 and 8.2) separations. Cellulose acetate electrophoresis enables provisional identification of Hb variants. However, many bands reflecting different Hb variants overlap (such as the band for HbS overlaps the band for HbD). The use of citrate agar electrophoresis (separates HbC from HbE) and knowledge of patients ethnic background (HbC is common in North Africa and HbE in South East Asia) improves interpretation of results. Quantification by densitometry is possible but not routine. Variants such as HbS can be quantified but this method is not accurate at a low percentage of abnormal Hb or for HbA2 quantification. HbA2 quantification by capillary zone electrophoresis (CE) and CE with isoelectric focusing (IEF, see below) has also been described. Separation of haemoglobins by electrophoresis is based on the relative charge of the αβ dimer and hence mutations that do not alter the charge may be “electrophoretically silent”. Electrophoresis is not a good detection method for fast moving variants such as HbH. Overall, electrophoresis methods are slow, insensitive and limited in versatility.
In aqueous solutions, a pH can be obtained by titration methods at which the net charge of a specific polypeptide or an amino acid is zero. This is the isoelectric point or pl. Isoelectric focusing is a polypeptide separation technique based on exploiting differences in pl values. Separation of Hb variants with similar charge has been achieved. It generates better resolution than electrophoresis. IEF has replaced the conventional electrophoretic methods used in many laboratories and has been used to identify a few Hb variants. Separation of polypeptides is achieved using a set of synthetic ampholytes with pl values that cover the range of the pls of the polypeptides to be separated, and a separation can be achieved with a pl difference of about 0.01 pH units on a support matrix. Even higher resolution is achieved with a pl difference of 0.001, if the ampholytes are bound to the matrix. The most commonly used IEF technique, not compatible with automation, is the application of multiple samples to a commercially prepared thin layer gel. IEF has the same limitations as electrophoresis methods. In common with electrophoresis methods, IEF methods provide no information on the molecular structure of the Hb.
Ionic and hydrophobic interactions of the sample with the supporting matrix are the basis of separation in ion exchange (IEX) and reversed-phase high-performance liquid chromatography (RP-HPLC) respectively. Hb can be isolated as an intact tetramer or the individual globin chains can be separated. HPLC has been used in the analysis of HbA2 HbF, other Hb fractions in screening for thalassaemias, as well as the isolation, detection and characterisation of several other Hb variants. Cation exchange chromatography, automated pre-programmed cation exchange HPLC and reversed-phase HPLC are used in laboratories for presumptive identification of haemoglobinopathies and thalassaemias. For definitive diagnosis, it is necessary to however still necessary to perform a DNA analysis or amino acid sequencing. These methods are time consuming, and do not give detailed information on the molecular structure of the variant and cannot be readily employed for high throughput screening tasks.
The genetic approach for detection and confirmation of diagnosis is an alternative strategy to polypeptide-based techniques, most of which are presumptive, especially where a mutation causes production of an unstable Hb, The development of polymerase chain reaction (PCR) methodologies and nucleotide sequencing techniques allows Hb variant characterisation at the gene level. A variety of methodologies have been developed for the detection of point mutations or deletions of α and β globin chains using DNA derived from white blood cells, amniocytes or chorionic villous samples. Southern blot oligonucleotide hybridisation, endonuclease restriction enzyme cleavage analysis of PCR products, amplification refractory mutation system, Gap PCR of known mutations, denaturing gradient gel electrophoresis and direct sequencing for unknown mutations are commonly used techniques.
All of the above methods require as much as hours to days to complete analysis and obtain the final result and are technically complex procedures. Recently, a prenatal real time PCR diagnostic method using the LightCycler requiring less than three hours including DNA extraction from a foetal sample (when parental mutations are known) has been described. Although DNA analysis is a powerful tool for identifying mutations or deletions, known and unknown, it cannot identify post-translational modifications of the expressed haemoglobins, and can only retrospectively give information about the origin of such changes.
For the analysis of polypeptides such as Hb variants, complete sequence coverage in a particular mass/charge window rather than a complete digest is preferred, in order not to lose the fragments smaller than 500 Da. This may be achieved by controlled incomplete proteolytic digest yielding overlapping fragments. In the Examples below, deliberate and controlled incomplete tryptic digestion of Hb in blood was performed to obtain analysable fragments to achieve a high level of sequence coverage, as compared to complete proteolytic digestion. The smaller fragments or fragments which are known to precipitate or those which are difficult to detect, as for example αT12 αT13, βT10, βT12, were consequently captured, since they are joined to bigger, more soluble fragments. A 100% sequence coverage for both the α and the β chain was achieved with trypsin using this newly developed digest method. The results were reproducible even after 6 months of sample storage.
The digestion may be carried out in solution or in an on carrier mode. It has been found that an on carrier digest provides superior performance. An on carrier digest typically digest includes the following steps, 1 μl of sample is deposited on 2 μl air dried trypsin on a MALDI sample plate, incubated for catalysis, stopped, covered with matrix and analysed. A detailed time course investigation has revealed the identity of fragments produced and the overall sequence coverage obtained for a particular time point. This procedure has dramatically improved sequence coverage, decreased digest time and robustness of the digestion chemistry. The data show that the acid labile surfactant sodium 3-[(2-methyl-2-undecyl-1,3-dioxolan-4-yl)methoxy]-1-propanesulfonate considerably reduced the digestion time of Hb when used with unpurified whole EDTA-treated blood. In combination with an on carrier digest, and the use of this surfactant, a 100% sequence coverage could be obtained for both globin chains in the a digest time of 2-3 min. This sulfonate-based surfactant with a monoisotopic mass of 417.2281 Da is acid labile and degrades to two non-interfering by-products with masses of 238.0482 and 198.1978. Such degradability has been reported for other sulfonate surfactants used with MALDI-TOF MS. It has been recognized that buffer components and surfactants, impose MALDI-ToF MS compatibility problems in terms of ionisation suppression. The development of sodium 3-[(2-methyl-2-undecyl-1,3-dioxolan-4-yl)methoxy]-1-propane-sulfonate and other acid-labile surfactants, which can be actively degraded to non-interfering by-products, show a new adaptation and streamlining of chemicals and methods in proteomics.
Whilst investigating variation of sample concentration with dilutions 1:10 and 1:100 with sodium 3-[(2-methyl-2-undecyl-1,3-dioxolan-4-yl)methoxy]-1-propanesulfonate and a 3 min digest, since the ionic surfactant concentration was equal in both digests, it can be concluded that in the 1:100 dilution digest the incompleteness of the digest is achieved not due to a lack of surfactant, but rather due to its intrinsic properties. The surfactant may only be able to interact by disintegration of the proteins on the domain-domain or the tertiary structural level. This indicates that an additional robustness level can be achieved with invariance of the sequence coverage in relation to the blood concentration. The computational analysis of the spectra of other blood proteins within the 25 highest MOWSE scores show that for each of the two dilution levels different proteins were identified by the Protein Prospector software. Further experiments and data analysis is essential to identify blood signature peptides other than those described from Hb for a particular dilution.
Besides the use of trypsin for on carrier 3 min digest in presence of sodium 3-[(2-methyl-2-undecyl-1,3-dioxolan-4-yl)methoxy]-1-propanesulfonate at 37° C. endoproteinase Glu C was investigated with success. The fragments produced by an on carrier Glu C digest with the particular conditions used in this invention enhance the overall peptide mapping capability
High quality mass spectra were obtained using automated data acquisition with set criteria. Rapid data acquisition with high resolution and signal to noise ratio was achieved without failure resulting in a high number of proteolytic peptides being identified within 10 ppm mass window. To test the robustness of the proteolytic method, various trypsin to sample ratios were investigated for on carrier 3 min digestion with sodium 3-[(2-methyl-2-undecyl-1,3-dioxolan-4-yl)methoxy]-1-propanesulfonate. The results show that varying the trypsin concentration from 5.45 pM/μl to 0.05 pM/μl did not alter the proteolytic fragmentation patterns adding to the robustness of the method.
Appearances of tryptic autolytic fragments have been reported in the literature. In this invention, a few autocatalytic fragments of trypsin were seen but to a much lesser extent than reported previously. This was most likely because of the short digest time due to the use of sodium 3-[(2-methyl-2-undecyl-1,3-dioxolan-4-yl)methoxy]-1-propanesulfonate and the high abundance of Hb in blood. The appearance of the autolytic fragments was sample to trypsin ratio dependent whereby decreased trypsin concentration or increased sample amount decreased the appearance of tryptic autolytic fragments. The autolytic fragments are thus not suitable candidates for internal calibration in the newly established method.
The methods developed have been used to identify a number of Hb variants. A total of 11 different α and β chain variants were identified by this method (Tables 1 and 2).
MVPLTP
(K/V)EK
Overall, the results demonstrate the general applicability of the 3 min on carrier proteolytic digest in the presence of the novel surfactant sodium 3-[(2-methyl-2-undecyl-1,3-dioxolan-4-yl)methoxy]-1-propanesulfonate at 37° C. for the identification of Hb variants.
The consistency in mass accuracy achieved by MALDI-TOF MS was remarkable (±5 Da) for intact globin chain analysis of Hb variants using this methodology. The intact globin chains, the matrix adducts and glycated globin chain adducts were well separated. The present applicants found that a better peak separation for globin chains, whilst resolving a variant heterozygous state, was achieved with a grid voltage of 90% and a delay time of 350 ns. It was evident from the spectra obtained in the linear mode for the variants observed that, although a high mass accuracy was achieved, a mass shift of <5 dalton cannot be identified with confidence, with the current specification of the MALDI-TOF MS instrument that was available. As such, whilst a protein identification can be established with a 10 ppm mass accuracy of any of its peptides greater than 11 amino acid residues in size from a MALDI-TOF MS spectrum in the reflector mode, the unambiguous proof of the absence of protein mutations requires both the determination of the mass of the protein in the linear mode and the complete coverage of the sequence obtained from proteolytic peptide mapping.
It was also observed that the peak area and relative intensity for the a and βchain was consistent for an individual sample and was highly reproducible for the same sample. The peak intensity and peak area for the α chain was persistently higher than for the β chain with a consistent α:β ratio in agreement with reports in the literature.
It was demonstrated that MALDI-TOF MS measurements to quantify Hb chain levels were possible by measuring the peak areas, although low abundance haemoglobins (<1%) cannot be quantified with current instrument settings. The quantitative utilities of MALDI-TOF MS have been reported in the literature. Analysis of the heterozygous state of the Hb S and sickle thalassaemia to quantify respective haemoglobin levels reflected similarity of results obtained with HPLC.
Glycated haemoglobin chains were also investigated to evaluate the quantitative aspects even further. It was observed that both the α and the β chain were glycated. It was also demonstrated in the experiments that glycation level was higher in the β chain than in the α chain. It was noted that there was a clear elevation of the glycated haemoglobin percentage in diabetic patient samples in agreement with reports in the literature. The MALDI-ToF MS measurements of glycated α and β chains resulted in a slightly higher percentage than reports obtained by a HPLC method, which only measures HB A1C (β chain only), whilst MALDI-TOF MS measurement was calculated using the whole pool of glycated globin chains. The MALDI-ToF MS measurements of only the glycated β chain were closer to the results obtained by an HPLC method, although it was observed that the MALDI-ToF MS measurements of glycated β chain were always lower than that of HPLC. Similar finding have been reported by Lapolla et al. In contrast to Lapolla, no globin chain preparation was employed and SA adducts were separated which was not reported by these authors. Furthermore, in contrast to Lapolla, the MALDI-ToF mass spectra obtained resolved the α, β and the glycated globin chains with a mass accuracy of 1.5 Da. Repeated testing resulted in the remarkable reproducibility of the area measurement (SD 0.01%). It was interesting to see that the sample with a HPLC A1C of 8.8% gave a higher MALDI-ToF MS measurement (14.71%) than the sample with a HPLC A1C of 10.0%. It was noteworthy that the glycated p globin chain MALDI-ToF MS measurement for both the samples were near to the results obtained by HPLC, where by the sample with HPLC reported percentage of 8.8% had a high a chain glycation.
Whilst investigating the two identified p globin glycated peptides derived by an on carrier 3 min endoproteinase Glu C digest in the presence of sodium 3-[(2-methyl-2-undecyl-1,3-dioxolan-4-yl)-methoxy]-1-propanesulfonate, it was observed that only the proteolytic derived glycated fragment βG8 Gluc-Hydr show an increased ratio for the glycated sample for peak areas, peak heights and relative intensities when compared with the respective values from the adjacent peak βG4-5. The proteolytic peptide fragment βG3-4 Gluc-Meth did not show any difference between the normal and the sample with high glycation level. The glycation of peptides may affect tryptic fragmentation pattern by blocking particular cleavage sites and the mass spectra may contain new glycated peaks.
If one uses the direct analysis approach, whereby the sample is merely diluted, then the relative proportion of a particular Hb chain (γ, δ, ζ) in relation to the α chain and β chain remains constant. This is a definite advantage if quantitation is the aim. In attempting low abundance detection of the ζ chain peptide fragments, the aim was not to achieve a particular sequence coverage, or detect ζ chain variants. The aim was to find the detection conditions, where pathological levels of the ζ chain could be detected in relation to normal globin chains levels.
In determining detectability of proteolytic peptides from digests of low abundance proteins, it was demonstrated that tryptic fragments of both the α and the β chain can be detected from digests performed with a 1:100000 dilution of whole human blood without purification. The low abundances of δ and γchains make the peptides derived from enzymatic digests of these chains difficult to detect, yet in this study, the detectability of the ζ chain in blood samples from patients with α thalassaemia was investigated. Huisman et al. reported elevation of ζ chain level in α globin gene mutations. The presence of embryonic ζ chain in adults has been used as a marker of the presence of a thalassaemia, and an ELISA method has been reported to detect the embryonic ζ chain in α thalassaemic individuals. Three different dilutions of blood samples, three from patients having a --/αα (-α3.7/-α3.7, -α3.7/--SEA) gene deletion and one normal haemoglobin from blood of a healthy individual, 1:10, 1:100 and 1:1000 with ammonium bicarbonate buffer, were investigated with successful identification of the ζT3 and the ζT5 in all three samples in all dilutions when 50 spectra were accumulated. The mass accuracy of the identified ζ chain fragments was low which is expected because of the extremely low abundance of the ζ chain fragment ions. The presence of the ζT3 and the ζT5 in all three dilutions and the absence of any ζ tryptic fragments in the normal blood sample spectra established MALDI-ToF MS as a potential screening tool for two gene deletion α thalassaemia, where an elevation of ζ chain levels is reported.
Thus, as discussed above, the present invention provides improved methods for polypeptide analysis. Particular applications of these new methods include the analysis of polypeptide variants. The present invention therefore provides for the use of these methods in the analysis of polypeptide variants. Also provided by the present invention are methods of diagnosis incorporating the methods of the present invention.
Various embodiments of the present invention will now be discussed by reference to the following non-limiting examples. While these examples focus on haemoglobin analysis, it is to be understood that the use of haemoglobin is illustrative and not to be taken as limiting the invention in any way. Haemoglobin has been chosen as it represents a class of polypeptides which demonstrates many well characterised variants. Furthermore, the usefulness of techniques of the present invention can be demonstrated to clearly discriminate between these many variants. The skilled addressee will recognise the applicability of these techniques to other polypeptides.
Throughout the specification and examples the following abbreviations are used.
ACN Acetonitrile
CHCA α-Cyano-4-hydroxycinnamic acid
CID Collision-induced dissociation
DE Delayed extraction
EDTA Ethylenediamine-N,N,N′,N′-tetraacetic acid
ELISA Enzyme-linked immunoassay
ESI Electrospray lonisation
Da Dalton
DHB Dihydroxybenzoic acid
DNA Deoxyribonucleic acid
Hb Haemoglobin
HPLC High-performance liquid chromatography
IEF Iso-electric focusing
LC Liquid chromatography
MALDI Matrix-assisted laser desorption/ionisation
min Minutes
MOWSE Molecular weight search
MS Mass spectrometry
m/z Mass-to-charge ratio
PSD Post source decay
SA Sinapinic acid
s Seconds
ToF Time-of-flight
TFA Trifluoroacetic acid
αCHCA α-Cyano-4-hydroxycinnamic acid
ppm Parts per million
Whole human blood samples, Hb standard and all the proteolytic digest products were analysed with a Voyager DE-STR MALDI-TOF mass spectrometer from Applied Biosystems, Framingham, Mass., U.S.A. The instrument was chosen because it has the highest mass accuracy amongst currently available MALDI-TOF instruments. The system uses a 337 nm nitrogen laser using 3-nanosecond duration pulses with a maximum firing rate of 20 Hz. The mass analyser is equipped with the Voyager DE-STR Biospectrometry Workstation software. All samples were spotted on 100 well stainless steel plates. A Perkin Elmer Cetus DNA thermal cycler from Narwal, U.S.A. was used for sample incubation and as a hot plate. A hot air oven from Watson Victor Ltd, Australia and water baths from Grant Instruments (Cambridge) Ltd, Cambridge, U.K. were used for incubation of sample plate and samples respectively: The balance used for measuring reagents was from Eppendorf, Netherler Hinz GmbH, Germany (Mettler Toledo AG245), the centrifuge (Biofuge B) from Heraeus Christ, Germany and the pH meter (pH 20) from ATI Orion Research, U.S.A. To measure the glycated Hb percentages high performance liquid chromatography with the TOSOH Glycohaemoglobin analyser HLC-723 GHbV A1c 2.2, Japan was used.
Human Hb A standard [9008-02-0] as well as proteins and peptides used as calibration standards, ie, angiotensin 1, ACTH (1-17), ACTH (18-39), ACTH (7-38), bovine insulin, thioredoxin (E. coli), equine apomyoglobin, were obtained from Sigma Chem. Co. (St Louis, Mo., U.S.A) to be used as calibration standards. The calibration standards were dissolved in ACN:H2O (50:50) (dilution) (v:v), 0.1% TFA. Proteolytic enzymes, bovine trypsin (10000 BAEE units/mg) [9002-07-7], endoproteinase Glu C [66676-43-5] and endoproteinase Asp N [9001-92-7] were obtained from Sigma Chem. Co. (St Louis, Mo., U.S.A). Ammonium bicarbonate and calcium chloride were obtained from BDH Chemicals (Kilsyth, Australia). Matrices α-Cyano-4-hydroxycinnamic acid (CHCA) [28166-41-8], 3,5-dimethoxycinnamic acid (Sinapinic acid, SA) [530-59-6] were obtained from Agilent (Forest Hill, Victoria, Australia) and 2,5-Dihydroxybenzoic acid [490-79-9] from Sigma Chem. Co. (St Louis, Mo., U.S.A). RapiGesT™ SF [308818-13-5], the ionic surfactant sodium 3-[(2-methyl-2-undecyl-1,3-dioxolan-4yl)methoxy]-1-propanesulfonate, was obtained from Waters (Rydalmere, NSW, Australia). Acetonitrile [75-05-8] (HPLC grade) and methanol [67-56-1] (HPLC grade) were obtained from Biolab Scientific Pty Ltd (Sydney, Australia). Trifluoroacetic acid (TFA) was obtained from Auspep Pty Ltd (Melbourne, Victoria, Australia). Water used for the study was distilled and deionised in a Milli-Q water purification system (Millipore, Bedford, Mass., U.S.A.).
HPLC and DNA sequencing was performed using standard protocols at the Monash Medical Centre. The results were used to select a variety of Hb variants to build up a database of identifiable Hb aberrations with mass spectrometry.
Accessible surface area for the amino acids from the globin chains of human Hb was calculated from the 1A3N file identifier taken from the Brookhaven Protein Data Bank (PDB) available at http://www.rcsb.org/pdb/ that utilizes the SCRIβT1 program available at http://www.bork.embl-heidelberg.de/ASC/asc2.html. The monoisotopic mass differences were calculated using the following atomic masses of the most abundant isotope of the elements, C=12.0000000, H=1.0078250, N=14.0030740, O=15.9949146, P=30.9737634 and the average masses were calculated using the following atomic weights of the elements C=12.011, H=1.00794, N=14.00674, O=15.9994, P=30.97376, S=32.066.
The numbering system of the sequence position used to describe the peptide fragments derived from the digests is the common protein-based description. In this system the amino acid after the initiator methionine is number 1 and the tryptic, endoproteinase Glu C and endoproteinase Asp N fragments are numbered according their occurrence in the amino acid sequence starting from the N-terminus.
Different instrument settings were systematically investigated to for high quality data acquisition.
Spectra were obtained with delayed extraction using a delay time of 250-350 ns, a grid voltage of 85% to 90%, with positive polarity. The mass range was 5000-100000 Dalton with a lower mass gate set at 5000 Da for mass data acquisition. Each spectrum was obtained with 500 laser shots by accumulating 5 spectra each obtained by 100 laser shots. Otherwise, automated spectra acquisition was used to collect 10 spectra, each spectrum obtained by 100 laser shots, using defined selection criterion for each spectra. Each spectrum was accumulated when it passed the selection criteria of minimum resolution of 200, 300 or 500, a minimum signal intensity of 10000, a maximum signal intensity of 64,000. The laser intensity was varied from 2500 to 3000. Central bias was used for automated data acquisition. 10 consecutive spectra without any selection criterion were accumulated using automated spectra acquisition for sample spectra failing to pass selection criteria. Manual acquisition was used for non-homogenous sample spots.
Spectra were obtained with delayed extraction using a delay time of 250 ns with positive polarity. The grid voltage was set at 85%. The mass acquisition mass range was 650-10000 Dalton where the low mass gate was set at 500 Da. Again, each spectrum was obtained with 100 laser shots and 5 consecutive spectra were accumulated. Automated spectra acquisition was used to collect either 10 or 50 spectra, each spectrum obtained by 100 laser shots, using defined selection criterion for each spectrum. Each spectrum was accumulated when it passed the selection criteria for selected peptides of a minimum resolution of 8000-10000, a minimum signal intensity of 1000 and a maximum signal intensity of 64000 for the base peak. The laser intensity was fixed to 2400 and central bias was used for automated data acquisition.
The resulting spectra were processed with the Data Explorer Software, Version 4.0.0.0, for baseline correction, noise filtering/smoothing and de-isotoping with the generic formula C6H5NO. Spectra were analysed using the ProteinProspector software ver. 3.2.1 using various settings to test automated identification of high and low abundance haemoglobins. For further analysis the 50 most intense peaks above a base peak intensity of 0%, 1% and 2% were considered. In this procedure the identity search mode was utilized were the IntelliCal routine utilises two filters for the obtained peaks list allowing for a maximum of five missed cleavage sites. Other setting for the procedures were requirement of a minimum of two peptides for a protein identification (considering the possibility of an acetylated N-terminus), allowing a protein molecular mass range from 1000-100000 Da, the pre-processing filter set to a mass accuracy of 150 ppm and the post-processing filter were set to a final mass accuracy of 10 ppm. For the automated detection of Hb ζ chain, the pre-processing filter was set to a mass accuracy of 400 ppm and the post-processing filter was set to a final mass accuracy of 250 ppm, the mass range to 5000-16500, and the pl range to 6.5-9. The peak filter was used to exclude the masses (m/z) below 650. This filtering was necessary as in Hb or blood digest, the heme group signal was overpowering the spectra most likely acting as an energy sink. The databank used for the identification of the Hb peptides was SwissProt mar03 and NCBInr.Mar03. Another search within the genepept 11299 databank was also conducted with the same settings. To automatically identify and label proteolytic fragments, a labelling file was created using the ‘create macro’ function of the Data Explorer Software, Version 4.0.0.0, containing the theoretical masses of ,
,
, γ, ζ globins, tryptic, Glu C and Asp N fragments up to five missed cleavages, their post-translational modifications and some possible artefacts masses. Peak area, ion count, peak resolution, peak height and peak relative intensity was calculated using the Data Explorer Software, Version 4.0.0.0.
Whole human blood samples collected in the haematology and the clinical genetics laboratories of Monash Medical Centre for electrophoresis, HPLC and DNA study were used. The samples were collected in EDTA (1.5±0.25 mg/ml of blood) containing vacutainers. These samples were then further subjected to mass spectrometric analysis. 5 μl of each of the blood samples was collected in eppendorf tubes from these laboratories and transferred, in iced containers. To investigate the stability of diluted whole blood in respect to MALDI-TOF MS analysis, blood samples diluted 1:100 in 50 mM ammonium bicarbonate buffer, 2 mM CaCl2, pH 8.3, stored in cold room and analysed at different time points. In order to trial a comparatively simple sample collection procedure with volumes smaller than 1 μl, 0.5 μl of blood was collected from two individuals using a pipette and blood was directly added to 50 mM ammonium bicarbonate buffer, 2 mM CaCl2, pH 8.3. The lysed blood was stored in a cold room for further analysis at different time points.
All samples were stored in a cold room at +4° C.
The only pre-MS sample preparation was dilution of blood. 1 μl whole human blood in EDTA, diluted 1:10, 1:100, 1:1000 and 1:10000 with buffer (50 mM ammonium bicarbonate buffer, 2 mM CaCl2, pH 8.3) or with deionised water for linear mode MALDI-TOF MS analysis of intact globin chains, adducts and post-translational modifications. To investigate the detectability and optimise the sample concentration in the reflector mode, samples were diluted 1:100, 1:500, 1:1000, 1:5000, 1:10000, 1:50000 and 1:100000 in ammonium bicarbonate buffer and proteolytic digestion was performed for each dilution in presence of a novel degradable surfactant.
Optimal sample preparation is a prerequisite for successful MALDI-ToF mass spectrometric analysis of peptide and protein samples. Variables associated with a good sample preparation to achieve high quality mass spectrometric data have been widely investigated for biological samples. In this invention, the sample preparation typically involves a dilution of whole human blood, which is the first step of the analysis of intact globin chains of haemoglobin [or of the proteolytic digestion products of the globin chains] and was systematically investigated. Anticoagulant EDTA-treated whole blood was used because this sample collection protocol is standard in clinical laboratories. Blood was investigated without any purification, and as such, no electrophoretic or chromatographic sample purification procedure was employed.
The amount of blood used in this investigation was 1 μl per sample. The samples were diluted and kept at 4° C. and subjected to experimental procedures at different time points.
Choice of matrices, sample matrix preparations and spotting methods are of utmost importance to achieve high resolution and high accuracy in mass measurements. Different sample spotting methods were investigated to achieve the desired resolution followed by further systematic investigations to improve and optimise each step of the Hb or Hb variant identification as described in the following sections.
Whole human blood in EDTA, diluted 1:10, 1:100, 1:1000 and 1:10000 with buffer (50 mM ammonium bicarbonate buffer, 2 mM CaCl2, pH 8.3) or with deionised water was spotted on the sample plate using different sample spotting methods described in the literature namely the two layer method, the sandwich method and the dried droplet method.
The samples were spotted with the two-layer technique by successive spotting 2 μl or 1 μl of either the matrix sinapinic acid (SA) or otherwise α-cyano-4-hydroxycinnamic acid (CHCA) and 1 μl of sample to have a matrix to sample ratio or 2:1 and 1:1 respectively.
For the dried droplet sample spotting method, 2 μl of sample and 2 μl of either the matrix SA or alternatively α-CHCA was mixed together, the sample-matrix mixture was further diluted 1:1, 1:5 and 1:10 with 50% ACN followed by deposition of 2 μl of this premixed sample matrix mixture on the sample plate for MS analysis.
For the sandwich sample spotting method, 1 μl of either the matrix SA or otherwise α-CHCA was spotted, air dried, followed by 2 μl of sample which was also air dried, which then followed by another 1 μl of either matrix on top of it.
A new sample spotting method was developed using a reversed-two-layer sample-spotting technique, whereby 1 μl whole human blood, diluted 1:10, 1:100, 1:1000 and 1:10000 with ammonium bicarbonate buffer, or with deionised water, was spotted on the sample plate, allowed to air dry, followed by addition of either 0.5 μl or 1 μl of SA. The in-solution tryptic digests were spotted using the reversed two layer method as well. The same reversed layer method was applied to analyse the on carrier digests in contrast to the commonly used method whereby matrix is directly added to the liquid analyte.
Variation in the sample-matrix crystallisation patterns with the different sampling methods was observed using diluted blood as the sample and SA as the matrix, as shown in
Whilst comparing (Table 3) different sample spotting methods, the dried droplet, the two-layer method the sandwich method and the new technique in this application, the new spotting technique gave the best results. The methods were compared in respect to signal to noise ratio, resolution, ion abundance and time taken to accumulate a defined number of spectra (5, 10 and 50) with set selection criteria. These significant modifications that have lead to the new sample spotting method have not previously been discovered. Although the specific case of Hb's have represented the model system to establish this new technique, it should be noted that the same methodology should be applicable to other proteins and their derived tryptic (enzymatic) fragments when they are analysed in the linear and reflector mode of MALDI-ToF mass spectrometric analysis.
The results demonstrate that the new spotting method described herein, whereby the diluted blood sample was spotted first, air dried, and then overlaid with the matrix (preferably sinapinic acid (SA)) using a sample matrix ratio of 2:1, substantially higher ion counts, higher resolution and excellent signal to noise ratios in the mass spectra were obtained both in the reflector and in the linear mode. This method gave a thin homogenous layer of sample matrix co-crystallisation, resulting in high spot-to-spot reproducibility with no obvious ‘hot’ [i.e. sample concentration non-homogeneity] spots. The fine microcrystalline coverage of the sample spot was best suited for an automated data acquisition whereby a 100% success rate was achieved for obtaining high ion counts (>10,000), high resolution (>500), high signal to noise ratio (>1 to 5000) and shorter acquisition time (˜90 s/1000 laser shot spectra) with spectra selection criteria set to a minimum signal intensity of 10,000, a maximum signal intensity to 64,000 and the minimum resolution set to 500. These criteria and outcomes are significant above previous experience described for the MALDI-ToF mass spectrometric analysis of tryptic peptides. The main advantages of the new spotting method against the previously used dried droplet sample spotting method was high spot-to-spot reproducibility, the requirement for less matrix and obviously the elimination of the step of premixing the sample with matrix.
In this sample handling/collection method, 0.5 μl samples were directly added to 49.5 μl of buffer (50 mM ammonium bicarbonate, 2 mM CaCl2, pH 8.3) with a resulting dilution factor of 1:100. The MALDI-TOF mass spectrometric analysis of the samples in the linear mode in the 7000-17000 m/z range show that the single charged [M+H]+ and double charged [M+2H]+ Hb A α and β chains were resolved with a high mass accuracy and with an inherent error less than 1 Da for single charged α and β chains, as depicted in Table 4. The corresponding MALDI-TOF mass spectrum is shown in
Although good spectra were obtained for the 1:100, 1:1000 and 1:10000 dilutions, the method was developed for the 1:100 dilution instead of the 1:1000 dilution, since this is a convenient dilution factor for other researchers, and because in the MALDI-TOF mass spectra of Hb tryptic peptides some peptides appeared to be have a low ion current abundance at the level of 1:1000 dilution. The low ion abundance may result in these peptides being resolved with a lower mass accuracy and thus be unsuited for automated data analysis. The trade-off at higher sample concentration is the appearance of peaks derived from other blood proteins. Although these additional peaks complicate to a minor extent the spectral data analysis, requiring extra care for interpreting the accumulated mass spectra, they do provide additional information since their presence was found to correlate with the conditions employed for the sample preparation, kinetics of enzyme digestion, digestion time, etc, thus enabling these non-Hb associated peaks to be used as “internal standards” for the detection other Hb chains within the sample with an abundance of >2% in relation to the α- and β-chains.
The concentrations of the unpurified blood samples were varied in order to optimise the selection of the dilution factor. Blood diluted with either with 50 mM ammonium bicarbonate buffer, 2 mM CaCl2, pH 8.3, or with deionised water gave similar results for all dilution factors when the reversed two layer sample spotting method using a sample to matrix ratio of 2:1 (sample 1 μl, matrix 0.5 μl) was employed. This outcome was not observed when other types of matrix compounds, such as α-CHCA and 2,5-DHB, were used. The 1 to 10 dilution of blood produced a non-homogenous sample spot. This also resulted in a very weak signal for both the α and β chain with no or a very poor separation of the matrix adducts of the chains, as shown in
Excellent MALDI ToF mass spectra were obtained for the 1:100, 1:1000, and 1:10,000 dilutions for the un-purified EDTA-treated blood in the linear mode, as shown in
The mass accuracy obtained for the dilutions 1:100, 1:1000, and 1:10,000 were persistently within 0.01%. The ion count was consistently above 10,000 for these dilutions, as shown in Table 7.
To find the best digestion conditions for human Hb α and β chains, and to assess the sequence coverage for both the chains and to document their proteolytic fragmentation pattern a time course proteolytic digest experiments on Hb A standard followed by whole EDTA treated diluted blood, normal Hb A and variant Hb E, were performed. Initially, in solution digests were performed followed by on carrier experiments to devise a rapid on carrier proteolytic digestion method with a novel degradable detergent. The optimised on carrier digest method was subsequently tested with some known and unknown variants, and with other proteolytic enzymes.
To optimise the digestion time and sequence coverage of the globin chains a time course experiment on Hb A standard was performed. 9 ml of the dissolved Hb A standard were incubated in a water bath at 37° C. for 5 minutes before adding 1 ml of a 10-fold trypsin stock solution. The final molar ratio of trypsin to Hb was 1:10. The 10 ml trypsin Hb solution was incubated at 37° C. in a water bath to allow the digestion process to occur. Aliquots of 250 μl were taken at time points 2, 4, 5, 6, 8, 10, 12 15, 20, 30, 45, 60 min and 2, 4, 8 and 24 hours and the digest was stopped with 62.5 μl 10% TFA (trifluoro acetic acid) yielding 83.7 μM Hb with a final concentration of 2% TFA for each time point. The samples were further diluted 1:5 with ACN:H2O (50:50) (v:v), 0.1% TFA for MS analysis. The samples were spotted with the two-layer technique by successive spotting 2 μl of either the SA or alternatively α-CHCA and 1 μl of sample. The final Hb concentration on the sample plate for each spot is 16.7 pmol/μl.
The first step in optimising the analysis of the whole human EDTA treated blood sample was to carry out a similar time course in solution tryptic digest experiment as for the Hb standard to document the fragmentation pattern and sequence coverage. In addition, applicability of the surfactant RapiGest™ (sodium 3-[(2-methyl-2-undecyl-1,3-dioxolan-4yl)methoxy]-1-propane-sulfonate) was investigated to enhance the efficiency of the proteolytic digest and to decrease the digest time. In this experiment, EDTA-treated whole human blood with an approximate Hb concentration of 9.3 mM (150 mg/mL) was diluted 1:100 (v/v) with 50 mM ammonium bicarbonate buffer, 2 mM CaCl2, pH 8.3. The diluted blood was subjected to a tryptic digest with and without a surfactant. For the digest without the surfactant 95 μl of blood and for the surfactant aided digest 90 μl of diluted blood was incubated with 5 μl 2% stock solution of the surfactant sodium 3-[(2-methyl-2-undecyl-1,3-dioxolan-4yl)methoxy]-1-propane-sulfonate in 50 mM ammonium bicarbonate buffer, 2 mM CaCl2, pH 8.3 at 37° C. in a water bath for 5 minutes. Then the digest was started by adding 5 μl of a 20-fold diluted trypsin stock solution (1.3 mg/ml) in 50 mM ammonium bicarbonate buffer, 2 mM CaCl2, pH 8.3 to both the samples to attain a final molar ratio of trypsin to Hb of 1:34. Both the samples were kept incubated at 37° C. to continue the digestion reaction and 10 μl aliquots were taken at different time points starting from 15 min and then 30 min, 1 h, 1 h 30 min, 2-8 hours in 1 h intervals and the last one at 24 hours. The digests were stopped by adding 2.5 μl 10% TFA to the aliquot of each time point yielding a final TFA concentration of 2%. For MS analysis the samples were then diluted 1:10 with ACN:H2O (50:50) (v:v), 0.1% TFA.
To optimise and develop a rapid, simple, robust proteolytic digestion method the following on carrier experiments were carried out using surfactant, initially with trypsin followed by independent experiments with endoproteinase Glu C and Asp N on whole normal blood and blood containing Hb variants.
2 μl of 20-fold trypsin stock solution with a trypsin concentration of 1.3 mg/ml (54.5 μM) equalling 5.45 pmole/μl was spotted for each digest on the sample plate and air dried at room temperature (22° C.) for 5 minutes. The sample plate was then incubated for 15 min at 37° C. and placed on a heating block or heating plate at 37° C. for 5 minutes before applying the sample. For on carrier tryptic digest of whole EDTA treated human blood, 19 μl of blood sample, diluted 1:100 with 50 mM ammonium bicarbonate buffer, 2 mM CaCl2, pH 8.3, was incubated either at 100° C. or at 37° C. for 5 min with 1 μl of 2% (w/v) of the surfactant sodium 3-[(2-methyl-2-undecyl-1,3-dioxolan-4yl)methoxy]-1-propane-sulfonate in 50 mM ammonium bicarbonate buffer, 2 mM CaCl2, pH 8.3. The sample incubated at 100° C. was cooled to 37° C. before adding the sample to the enzyme. 1 μl of this heat-denatured sample (93 μM Hb=93 pmole/μl) was spotted on the dried trypsin spots yielding a final molar ratio of trypsin to Hb for each spot of 1:17. The digestion reaction was stooped with 0.5 μl 10% TFA after 2 s, 10 s to 1 min at 10 s intervals, and then onwards to 3 min at 15 s intervals. Matrix, 0.5 μl of SA was added and the samples air-dried.
The optimised time for on carrier tryptic digestions in the presence of the surfactant sodium 3-[(2-methyl-2-undecyl-1,3-dioxolan-4yl)methoxy]-1-propane-sulfonate was then tested using the proteolytic enzyme Glu C. For the on carrier digestion of Hb in whole EDTA treated blood endoproteinase Glu C stock solution was made by dissolving 25 μg of lypophilized Glu C in 25 μl of 50 mM ammonium bicarbonate buffer, 2 mM CaCl2, pH 8.3. Then 1.5 μl of a further 50 fold diluted Glu C stock solution (1 μg/μl=34.5 μM) in 50 mM ammonium bicarbonate buffer, 2 mM CaCl2, pH 8.3 equalling 0.69 μM/μl was spotted for each digest on the sample plate and air dried at room temperature. 19 μl of blood sample, diluted 1:100 with 50 mM ammonium bicarbonate buffer, 2 mM CaCl2, pH 8.3, was incubated for 5 min with 1 μl of 2% (w/v) sodium 3-[(2-methyl-2-undecyl-1,3-dioxolan-4yl)methoxy]-1-propane-sulfonate in 50 mM ammonium bicarbonate buffer, 2 mM CaCl2, pH 8.3, at 37° C. After the heat denaturation step, before adding the sample to the enzyme, the sample plate was placed on a heating plate at 37° C. for 5 minutes. 1 μl of this blood sample was spotted on each of the dried Glu C spots yielding a final molar ratio of Glu C to Hb of 1:90 and stooped with 0.5 μl 10% TFA after 3 min. The samples were allowed to dry before 0.5 μl of SA was added.
The MALDI-TOF mass spectrum derived in the linear mode in the 5000-25000 m/z range for unpurified whole EDTA treated human blood containing Hb A (α2β2) show the double charged m/z values (received [M+2H]++/2: 7596.23 and 7959.33 (expected 7568.19 and 7934.61), the single charged m/z values (received [M+H]+: 15127.47 and 15868.31, expected 15868.23) and the m/z values for the α-,α-
,β-β dimers (received [M+H]+: 30173.07, 30914.66 and 31677.26) as shown in Fig. I. The m/z values of the single charged intact
chain and β chain of Hb A were measured with an error of 0.10 and 0.08 Dalton respectively. Errors associated with other peaks are listed in Table 8.
a6.61,
b5.04.
The masses of 15333.37 and 16078.54 with their respective mass differences of the received single charged α and β mass of 205.9 and 210.23 are considered to derive from Hb-SA adducts. Hb matrix adducts were also reported previously. The masses of 15292.81 and 16031.27 with their respective mass differences from the received single charged α and β mass of 165.3 and 163.0 are considered to derive from glycation of the respective chains, this finding is in agreement with previous reports. Errors associated with the peaks are listed in Table 9.
Initial experiments were designed to establish the time necessary to achieve a complete Hb standard digest followed by a time course experiment to document the sequence coverage of the respective globin chains at different time points using the enzyme trypsin. The sample procedure was that outlined in example 2.1.1. A complete digest was obtained after 24 hours, as judged by the disappearance of the Hb chains in the corresponding reversed-phase HPLC chromatograms (data not shown) and the MALDI-TOF mass spectra in the m/z range from 5000-25000 in the linear mode (data not shown). The time course of the free solution digests of the Hb A standard versus the sequence coverage is depicted in
The effect of the ionic surfactant sodium 3-[(2-methyl-2-undecyl-1,3-dioxolan-4yl)-methoxy]-1-propane-sulfonate (RapiGest™ SF) on the sequence coverage of the Hb A α- and β-chain in a free solution digest in 1:100 diluted EDTA treated blood was investigated by performing a time course experiment. The procedure followed was that of example 2.1.2. The results for the individual digest times in the absence and the presence of the surfactant are depicted in
With the surfactant RapiGest™ in a free solution digest, as shown in Panel B, a good sequence coverage was obtained for Hb A digestion times below two hours, with the excellent cleavage time of 15 min and a sequence coverage of 95.04% for the α chain and 82.19% for the β chain. Here, the fragments αT11 and βT13-15 were missing. Interestingly, at 420 min, the occurrence of βT10-13 coincides with the disappearance of αT12-14, as if these fragments would compete for ionisation and desorption. None of the fragments believed to precipitate, αT12, αT13, βT10 and βT12 could be detected. Although the occurrence of surfactant dimer and trimer formation with m/z values of 855.617 and 1271.922, respectively, was reported, only the dimer, with a m/z value of 855.617, was identified infrequently, under the conditions employed.
The sample was prepared according to the general procedure outlined in example 2.2.1. The sequence coverage of the Hb α and β chain in 1:100 diluted EDTA treated blood for an on carrier solution digest in the presence of the surfactant RapiGest™ SF after a 100° C. or 37° C. pre incubation is plotted in
For the on carrier digest at 37° C. in the presence of the surfactant, as shown in Panel D, the sequence coverage for both the chains was consistently high with a plateau between 90 s and 180 s. Detection of the αT12-14 explains the obtained 100% sequence coverage of the αchain within the plateau, which did not appear in shorter digestion times. For the β chain at each time point of the plateau, one fragment was missing, whereby the absence of the large βT12 fragment (16 amino acid) at 2 min had the highest impact, whilst all the other missing fragments were dipeptides, either βT6 or βT15. Complete sequence coverage was obtained for both, the α and the β chain, at 180 s. With these particular conditions, from 90-180 s, method robustness was achieved, i.e. where small changes in digestion time result in only small changes in sequence coverage. If the results from both on carrier experiments, the combined effects of heat plus surfactant denaturation for a pre-incubation at 100° C. and the surfactant denaturation alone (with the their plateaus from 10-60 s and 90-180 s, respectively) is analysed, it is obvious that the surfactant alone only partially denatures the proteins, whilst the additional heat increases the denaturation and thus the accessibility of additional cleavage sites.
The mass spectra corresponding to selected time points 10 s, 30 s, 90 s and 180 s in the on carrier digest at 37° C., are shown for the m/z range from 650-5600 in
To additionally monitor the digest from the disappearance of the intact globin chains, mass spectra corresponding to each time point were obtained in linear mode.
The spectrum in
All other Hb A tryptic fragments had a mass accuracy below 10 ppm and were not used in the computational identification procedure. In addition to the tryptic fragments of α and β globin chains of Hb A, 5 fragments of the δ-chain were identified. The δ-chain is homolog to the β-chain differing in 10 amino acids, one of which is, Arg116, resulting in an additional trypsin cleavage site. Since HbA2 (α2δ2) constitutes only less than 3% of the hemoglobins, the abundance of these peptides and consequently their mass accuracy was quite low, as shown in Table 12. Nevertheless, the method was considered able to detect aberrant high Hb abundances of HbA2 (α2δ2). At this stage and with the set conditions, no γ chain fragments from Hb F (α2γ2), present in very low abundance in normal human adult blood (<1%) were detected.
In the methods of the invention, autocatalytic tryptic fragments very rarely detected with low abundance or peak intensity, not surprisingly, firstly because of the shortness of the digest time, only 3 min, and secondly because of the inactivity of potentially present enzymes present (like serine-proteases), caused by the denaturing action of the surfactant. As the trypsin activity is maintained, it was anticipated that the trypsin concentration could be further decreased, leading to substantial cost savings in high-throughput applications. Moreover, the surfactant could increase the lifetime of the expensive enzyme-linked sample plates.
An on carrier digest, at 37° C., of two different dilutions of unpurified human blood with ammonium bicarbonate, 1:100 and 1:10, was performed with the presence of the ionic surfactant. The digests were stopped each at 2 min, and the obtained spectra of the digested blood sample for the two dilutions were compared. For the 1:100 dilutions, the sequence coverage for the α chain was 100% and for the β chain was 89.04% due the missing of βT12, as shown in
However, the thirteen Hb A (α2β2) fragments that were detected in the 1:10 dilution in the 10 ppm window, with increasing mass were, αT4, αT5, αT6, αT3-4, αT6-7, αT6-8, αT3-5, αT1-4, αT1-5, and βT4, βT3, βT2-3, βT1-3, as shown in Table 14, indicate that with increasing blood concentration the number of peaks detected with lower than 10 ppm mass accuracy increase.
The sequence coverage and fragmentation pattern of the Hb α and βchains in EDTA treated unpurified whole human blood, diluted 1:100 in ammonium bicarbonate, for an on carrier 3 min solution digest with endoproteinase Glu C in the presence of the surfactant RapiGest™ SF after 37° C. pre incubation was investigated. This followed the general procedure outlined in example 2.2.2. From a theoretical standpoint, a complete digest, which produces 5 α and 9 β fragments, as shown in Table 15 in the 650-5650 m/z window would correspond to a 34.04% sequence coverage for the α chain and a 88.36% sequence coverage for the β chain. The low sequence coverage for the α chain is due to the small number of fragments produced when subjected to endoproteinase Glu C digest, where out of the five possible fragments, two fragments are too small (αG2 and αG3) and one is too large (αG4) to be detected within the 650-5650 m/z window. For the β chain, there are more detectable fragments within the 650-5650 m/z window. For a 3 min on carrier digest, a sequence coverage of 21.28% for the α chain and 48.23% for the β chain was achieved, as shown in
The number of fragments detected within 10 ppm was 3 for each chain. The β chain of human Hb possess two consecutive endoproteinase Glu C specific amino acids, Glu6 and Glu7, and it was observed that endoproteinase Glu C hydrolysed the chain at both amino acid residues producing the fragments, in increasing m/z, βG1-2 (m/z value of 824.3936, pos 1-7), βG3 (m/z value of 1616.7608, position 8-22), βG2-3 (m/z value of 1745.9068, position 7-22), confirming the phenomena, as shown in
Since the best results for the on carrier tryptic digestion of Hb A in whole human blood were obtained with the ionic surfactant RapiGest™ SF at 37° C. at 3 min digest time, this procedure was applied to the blood samples with known Hb variants at a 1:100 dilution along with two samples with unknown Hb variants, listed in Table 16. When subjected to a digest, the Hb chain containing a substitution of an amino acid, due to the presence of a mutation in the corresponding gene, results in a mass shift of a specific fragment, the appearance of new signature peptide/s as a result of addition of a cleavage site or disappearance of fragment/s followed by appearance of new fragment/s as a result of deletion of a cleavage site. An elongation or a deletion of a chain segment would also be reflected by a mass shift of the corresponding fragment/s.
The monoisotopic masses of the peptide fragments were calculated with the program Peptide Mass at the ExPASy website http://kr.expasy.org/cgi-bin/peptide-mass.pl. The information on individual mutants was taken from the Globin Server at http://globin.cse.psu.edu.
In the following section the Hb variants are grouped according to the impact of the enzyme on the number of fragments and the use of enzymes.
Substitution of a particular amino acid with another amino acid which constitutes a cleavage site to a certain enzyme results in producing additional proteolytic fragments. The substitution of an amino acid with “Lys”, a specific amino acid for trypsin, in any position would result in two new tryptic fragments for a complete cleavage, resulting in an additional fragment. If incomplete cleavages occur, then several additional fragments may occur. These additional fragments can be used as signature peptide to identify Hb variants.
In the following, the newly developed method including a time course investigation is applied to Hb E. The Hb E (α2ββE) is characterised by a Glu26 to Lys26 mutation whereby the resulting βE chain differs from the normal β chain by a molecular mass of 0.94 Dalton. In Hb E, in one of the β chains, the normal βT3 fragment VNVDEVGGEALGR is converted to βET3 and βET4 by the introduction of an additional cleavage site VNVDEVGGK/ALGR, yielding two unique fragments with expected monoisotopic masses of [M+H]+ 916.4734 and 416.2616. As a consequence, all subsequent fragments of the βE-chain have to be renumbered, although they are identical, i.e. βT10=βET11.
The mass spectrum of human blood containing a Hb E (α2ββE) variant shows the double charged (received m/z values [M+2H]++/2: 7557.4 and 7927.8) and single charged (received m/z values [M+H]+: 15125.1 and 15869.4) Hb E α chain and β chains, respectively, whereby the βchain and βE chain, could not be resolved, as shown in
In this experiment, a time course on carrier tryptic digestion was performed. The on carrier tryptic digestion of Hb E in whole human blood obtained with the ionic surfactant RapiGest™ SF at 37° C. with a 3 min digest time resulted in a spectrum with 100% sequence coverage for the α chain and β chain, respectively, shown in
Overall the results demonstrate the general applicability of the newly developed method. In the following further experiments, no time course experiment was done; instead the optimised conditions for a 3 min on carrier digest in the presence of the novel surfactant at 37° C. were applied.
The Hb variant Hb C (α2ββC) is characterised by a Glu6 to Lys6 substitution, whereby the resulting βC chain differs from the normal βchain by a molecular mass of −0.94 Dalton. In Hb C, in one of the β chains, the normal βT1 fragment VHLTPEEK is converted to βCT1 and βCT2 by the introduction of an additional cleavage site VHLTPK/EK, yielding two unique fragments with expected monoisotopic masses of [M+H]+ 694.4246 and 276.1554. As a consequence, all subsequent fragments of the βC chain have to be renumbered, although they are identical, ie. βT2=βCT3.
The mass spectrum of human blood containing an Hb C (α2ββC) variant shows the double charged (received m/z values [M+2H]++/2: 7627.23 and 7994.77) and single charged (received m/z values [M+H]+: 15127.83 and 15868.13) Hb C α chain, β and βC chains, respectively, whereby the β chain and βC chain, could not be resolved, as shown in
The Hb C signature peptides βCT1 and βCT2 could not be detected with the current settings, as these smaller fragments were lost in the matrix background. However, a signature peptide clearly specific for the Hb C variant, βCT2-3, EKSAVTALWGK, was detected with a mass accuracy of 7.14 ppm (expected m/z value 1189.6575, received, 1189.6490) and thus the Hb C variant was identified, as shown in
A minor peak of a second Hb C signature peptide βCT1-2, VHLTPK/EK, was detected with a lower mass accuracy of −13.24 ppm (expected m/z value 951.5622, received 951.5748), which indicates that a 0.935 Da mass shift to the left can be detected with the settings used in this invention using the reflector mode, as shown in
The presence of the signature peptides for Hb C confirms its presence, but at the same time the presence of the βT1 [M+H]+ 952.4958 fragment is of high significance. The presence of this peak confirms the heterozygous state for Hb C and the presence of the normal β chain, whereby the absence of which would imply a homozygous state for the variant. The additional cleavage site may account for the low abundance of the βCT1-2 peptide in the digestion products. Since in a heterozygous state for haemoglobin C, only 30-40% of the total haemoglobin content is haemoglobin C, the decreased signal intensity of βCT1-2 (resolved m/z 951.5748) when compared with its normal counterpart can be explained. The low ion abundance for βCT1-2 may also be the reason for its low mass accuracy.
The haemoglobin variant Hb S (α2 ββS) is characterised by a Glu124 to Val124 (E to V) mutation in the β chain, whereby the resulting βS chain differs from the normal α chain by a molecular mass of −29.98 Da.
The mass spectrum of human blood containing a Hb S (α2ββS) variant shows the single charged [M+H]+ average m/z value of 15127.35 (expected m/z value 15127.37) representative for the α chain and the [M+H]+ average m/z values 15867.45 (expected m/z value 15868.23) and 15839.18 (expected m/z value 15838.25) for the β and βS chain, respectively, whereby the β chain and βS chain, had a mass difference of −30.3 Da (expected mass shift −29.98 Da), as shown in
In Hb S heterozygotes, due to substitution of an amino acid in one of the β chains, the normal βT1 fragment, [M+H]+ with a monoisotopic mass 952.5098, VHLTPEEK is converted to smaller tryptic fragments βST1, VHLTPEVK, with an expected monoisotopic masses of [M+H]+ 922.5356, the βT1-2 fragment, VHLTPEEKSAVTALWGK, [M+H]+ 1866.0119, is converted to βST1-2, VHLTPEVKSAVTALWGK, with an expected monoisotopic masses of [M+H]+ 1836.0377, and the βT1-3 fragment, VHLTPEEKSAVTALWGKVNVDEVGGEALGR, [M+H]+ 3161.6589, is converted to βST1-3, VHLTPEVKSAVTALWGKVNVDEVGGEALGR, with an expected monoisotopic mass of [M+H]+ 3131.6847. The on carrier tryptic digestion of Hb S heterozygote (α2ββS) in whole human blood obtained with the ionic surfactant RapiGest™ SF at 37° C. and 3 min digest time, yielded two signature peptides, the βST1, as shown in
The Hb variant Hb J-Bangkok, also known as Hb J-Korat, Hb J-Manado or Hb J-Meinung, (α2 ββJ-Bangkok) is characterised by a Gly56 to Asp56 (G to D) mutation in the β chain, whereby the resulting PJ-Bangkok chain differs from the normal β chain by a molecular mass of 58 Da.
The associated error was −0.29 Da for the chain, −0.99 Da for the β chain and 1.04 Da for the βJ-Bangkok chain. The split in the β chain confirms the heterozygous state. The mass spectrum of human blood containing an Hb J-Bangkok (α2ββJ-Bangkok) variant showed also the double charged globin chains m/z value [M+2H]++/2: 7605.08 and a split in second peak, 7974.37/8003.14), also shown in
In Hb J-Bangkok heterozygotes, due to substitution of an amino acid in one of theβ chains, the normal βT5 fragment with the monoisotopic mass [M+H]+ of 2058.9477, FFESFGDLSTPDAVMGNPK is converted to the βJ-Bangkok T5 fragment, FFESFGDLSTPDAVMDNPK, with an expected monoisotopic masses of [M+H]+ 2116.9531. The on carrier tryptic digestion of haemoglobin J-Bangkok (α2ββJ-Bangkok) in whole human blood obtained with the ionic surfactant RapiGest™ SF at 37° C. and a 3 min digest time produced the signature peptide, βJ-Bangkok T5, with a mass accuracy of −3.12 ppm, where as the its counterpart, normal βT5 was detected with a mass accuracy of 9.23 ppm, as shown in
The haemoglobin variant Hb Setif is an α chain variant (ααSetifβ2). It is characterised by an Asp94 to Val94 (N to Y) substitution in the α chain, whereby the resulting αSetif chain differs from the normal a chain by a molecular mass of +48.09 Da.
The mass spectrum of human blood containing a Hb Setif (αSetifβ2) variant shows the single charged [M+H]+ average m/z value of 15128.69 (expected m/z value 15127.37 Da) for the α chain, a [M+H]+ average m/z value of 15172.56 Da (expected m/z value 15175.46) for αSetif and a [M+H]+ average m/z value of 15868.46 (expected m/z value 15868.23) for the β chain. The α chain and the αSetif chain, had a mass difference of 44.79 (expected mass shift 48.09) as shown in
In Hb Setif heterozygotes, due to substitution of an amino acid in one of the chains, the normal αT11 fragment with a monoisotopic [M+H]+ mass of 818.4406, VDPVNFK is converted to αSetifT11, VYPVNFK, with an expected monoisotopic mass of [M+H]+ 866.4770 Da, and a αT10-11 fragment, LRVDPVNFK, [M+H]+ 1087.6258 Da, is converted to αSetif T10-11, LRVYPVNFK, with an expected monoisotopic mass of [M+H]+ 1135.6622 Da. The on carrier tryptic digestion of the Hb α variant, Hb Setif (ααSetif β2), in whole human blood obtained with the ionic surfactant RapiGest™ SF at 37° C. and a 3 min digest time yielded two signature peptides, αSetif T11, as shown in
The Hb Ty Gard (α2βTyGard) is a β chain variant and is characterised by a Pro124 to Gly124 (P to G) mutation in the β chain, whereby the resulting βTyGard chain differs from the normal β chain by an average molecular mass of +31.01 Da.
The mass spectrum of human blood containing a TyGard (α2ββTyGard) variant shows the single charged [M+H]+ average m/z value of 15128.7 Da (expected m/z value 15127.37 Da) representative for the α chain and a [M+H]+ average m/z value of 15868.40 Da (expected m/z value 15868.23 Da) and 15898.70 Da (expected m/z value 15899.24 Da) for β and βTyGard chains, respectively, whereby the βchain and βTyGard chain, had a mass difference of 30.3 Da (expected mass shift 31.01 Da) as shown in
In Hb TyGard heterozygotes, due to substitution of an amino acid in one of the β chain, the normal βG9 fragment with a monoisotopic [M+H]+ m/z value of 2680.4357 Da, is converted to βTyGardG9 with an expected monoisotopic mass [M+H]+ of 2711.4457 Da, as shown in Table 20.
The on carrier endoproteinase Glu C digestion of haemoglobin TyGard (α2ββTyGard) in whole human blood obtained with the ionic surfactant RapiGest™ SF at 37° C. and a 3 min digest time resulted in a spectrum with similar sequence coverage for the α chain and β chain, respectively, achieved for normal blood showing similar fragmentation pattern when digested with endoproteinase GluC, as shown in
The Hb variant Hb J Toronto (ααJ-Toronto β2) is characterised by an Ala5 to Asp5 (A to N) substitution in the α chain, whereby the resulting αJ-Toronto chain differs from the normal α-chain by a molecular mass of +44 Da.
The mass spectrum of human blood containing a Hb J-Toronto (ααJ-Toronto β2) variant shows the single charged [M+H]+ average m/z value of 15128.89 Da (expected m/z value 15127.37 Da) representative for the chain, a [M+H]+ average m/z value of 15170.19 Da (expected m/z value 15171.38 Da) for αJ-Toronto and a [M+H]+ average m/z value of 15868.84 Da (expected m/z value 15868.23 Da) for the βchain. The αchain and αJ-Toronto chain had a mass difference of 43.0 Da (expected mass shift 44.1 Da) as shown in
In Hb J Toronto heterozygotes the substitution of Ala5 to Asp5 (A to N) in one of th chain yields three signature peptides identifiable by a 3 min on carrier endoproteinase Glu C digest with RapiGest™ SF at 37° C. The first signature peptide is αJ-TorontoG1, VLSPNDKTNVKAAWGKVGAHAGE, with an expected mono-isotopic mass of [M+H]+ 2350.2149 Da, where as its counterpart, the normal αG1 fragment has a monoisotopic [M+H]+ m/z value of 2306.3896 Da (VLSPADKTNVKAAWGKVGAHAGE). The second signature peptide is a result of substitution in the αG1-2 fragment with 1 missed cleavage, VLSPADKTNVKAAWGKVGAHAGEYGAE, having a monoisotopic [M+H]+ m/z value of 2726.3896 Da. The αJ-TorontoG1-2 fragment, the second signature peptide, VLSPNDKTNVKAAWGKVGAHAGEYGAE, has an expected monoisotopic mass of [M+H]+ 2770.3794. The third signature peptide is converted from the normal αG1-2 fragment, VLSPADKTNVKAAWGKVGAHAGEYGAEALE, with a monoisotopic [M+H]+ m/z value of 3039.5533 Da. The αJ-TorontoG1-3 signature peptide fragment has an expected monoisotopic mass of [M+H]+ 3083.5432 Da (VLSPNDKTNVKAAWGKVGAHAGEYGAEALE).
The 3 min on carrier tryptic digestion of the Hb α variant, J-Toronto (ααJ-Toronto β2), in whole human blood obtained with the ionic surfactant RapiGest™ SF at 37° C. resulted in three signature peptides, the αJ-TorontoG1, as shown in
The normal counterparts of these fragments, the αG1, the αG1-2 and the αG1-3, were also detected with mass accuracy of 50.5, −36.6 and −61.0 ppm, respectively. The appearance of an additional peak besides the normal α peak with a mass shift of +43.0 Da in the linear mode and the detection of the three signature peptides unique for the Hb J Toronto variant in reflector mode unambiguously identified the sample to come from an individual carrying Hb J-Toronto. The two peaks in the liner mode and the detection of the αG1, the αG1-2 and the αG1-3 fragments confirm the Hb J-Toronto carrier state.
The Hb variant Hb J-Kaohsiung, (α2ββJ-Kaohsiung) is characterised by a Lys59 to Thr59 (K to T) change in the β chain, whereby the resulting βJ-Kaohsiung chain differs from the normal βchain by a molecular mass of −27.07 Da. The substitution of Lys, an amino acid which is a specific cleavage site for trypsin, to Thr results in the loss of a cleavage site. As a consequence, βT5 and βT6 merge to form βJ-KaohsiungT5, with a mass shift of −27.07 Daltons, and subsequent fragments of the βJ-Kaohsiung have to be renumbered, although they are identical, i.e. βT7=βJ-KaohsiungT6.
The mass spectrum of human blood containing a J-Kaohsiung variant, (α2ββJ-Kaohsiung) shows the single charged [M+H]+ average m/z value of 15127.00 Da (expected m/z value 15127.37 Da) representative of the α chain and a [M+H]+ average m/z value of 15867.80 Da (expected m/z value 15868.23 Da) and 15842.85 Da (expected m/z value 15841.16 Da) for the β and the βJ-Kaohsiung chains, respectively, whereby the βchain and βJ-Kaohsiung chain, had a mass difference of −25.55 Da (expected mass shift −27.07 Da) as shown in
In Hb J-Kaohsiung heterozygotes, due to substitution of Lys to Thr in one of the β-chains resulting in a deletion of a cleavage site, the normal βT5-6 fragment with a monoisotopic [M+H]+ m/z value of 2486.1110 Da, is converted to βJ-KaohsiungT5 with an expected monoisotopic mass of [M+H]+ 2259.0638 Da, the normal βT5-7 fragment, [M+H]+ 2679.3235 Da, is converted to βJ-KaohsiungT5-6 with an expected monoisotopic mass of [M+H]+ 2652.2762 Da, as shown in Table 22.
The on carrier trypsin digestion of Hb J-Kaohsiung (α2ββJ-Kaohsiung) in whole human blood obtained with the ionic surfactant RapiGest™ SF at 37° C. and a 3 min digest time allowed the detection of the signature peptides, βJ-kaohsiungT5, FFESFGDLSTPDAVMGNPTVK, with a monoisotopic mass of 2259.4464 Da (expected [M+H]+ m/z value 2259.0638 Da) and a mass accuracy of −169.3 ppm and βJ-KaohsiungT5-6, FFESFGDLSTPDAVMGNPTVKAHGK, with a monoisotopic mass of 2652.6727 Da (expected [M+H]+ m/z 2652.2762 Da) and a mass accuracy of −49.4 ppm, as shown in
From the theoretical point of view, the βJ-KaohsiungT5 fragment with a monoisotopic [M+H]+ m/z of 2259.4464 Da, has a identification conflict with the γT62-82 fragment with a monoisotopic [M+H]+ m/z value of 2259.2812 Da. However the mass value received can be seen as to belong to βJ-Kaohsiung since the low abundance of Hb F (α2γ2) in adult blood can be assumed.
The appearance of βJ-KaohsiungT5-6 (AA 41-65) was an interesting observation, as the normal βT5-7 (AA 41-65) fragment was not detected in this invention, as documented, in
Although the signature peptides for J-Kaohsiung (α2ββJ-Kaohsiung) were detected with lower mass accuracy, believed to be result of the low abundance of the peptides, appearance of two signature peptides unambiguously identified the Hb variant J-Kaohsiung (α2β2βJ-Kaohsiung). The appearance of two β peaks in the linear mode confirms the heterozygous state for a Hb variant J-Kaohsiung.
The Hb variant Hb Long Island, also known as Hb Marseille, (α2ββLongIsland) is characterised by an extension of the N-terminus by a Met (M) residue, and a His2 (H) to Pro2 (H to P) substitution in the β chain, whereby the resulting βLongIsland chain differs from the normal βchain by a molecular mass of 91.17 Da (Met addition would result in a 131.04 Da shift, the H is >Pro would result in a −40.2 Da shift, finally resulting in a net change of 131.04-40.02=91.17 Da).
The mass spectrum of human blood containing a Long Island (α2ββLongIsland) variant shows the single charged [M+H]+ average m/z value of 15127.47 Da (expected m/z value 15127.37 Da) representative for the chain and a [M+H]+ average m/z value of 15867.04 Da (expected m/z value 15868.23 Da) and 15957.86 Da (expected m/z value 15959.40 Da) for β and βLondIsland chains, respectively, whereby the βchain and βLondIsland chain, had a mass difference of 90.9 Da (expected mass shift 91.17 Da) as shown in
chain, −1.19 Da for the βchain and 1.54 Da for the βLongIsland chain. The mass spectrum of human blood containing a Hb LongIsland (α2ββLondIsland) variant shows the double charged [M+2H]++/2 m/z values of 7554.22 Da and a split of the second peak with m/z values of 7927.8 Da and 7938.96 Da.
In Hb LongIsland heterozygotes, one of the β chains, the normal βG1-3 fragment, [M+H]+ 2422.264, is converted to βLongIslandG1-3 with an expected monoisotopic mass of [M+H]+ 2513.10189 Da, as shown in Table 23.
The on carrier endoproteinase Glu C digestion of haemoglobin Long Island (α2ββLongIsland) in whole human blood obtained with the ionic surfactant RapiGest™ SF at 37° C. with a 3 min digest time resulted in a spectrum with similar sequence coverage for the α chain and β chain, respectively, achieved for normal blood showing similar fragmentation pattern when digested with endoproteinase Glu C with an extra peak, as shown in
The quantitative aspects of MALDI-ToF MS have been reported in the literature. In this invention quantitative aspects of MALDI-TOF MS in respect to haemoglobinopathies have been explored. Variation of different Hb levels is characteristic of many β Hb variants. The following table (Table 24) represents the level of different Hbs characteristic for some β thalassaemias and their interactions with Hb variants (modified).
Four patient samples from known sickle thalassaemia and Hb S heterozygote with known HPLC results were investigated using the MALDI-ToF MS linear mode. The peak area represents the ion species abundance which reflect the amount of the proteins. The peak area was calculated using the Data Explorer Software and the sum of the peak areas representing , βs, δ and γ chains were added (100%) proportions were calculated accordingly. For each sample, 5 consecutive spectra were obtained whereby each spectrum was an accumulation of 5 spectra each obtained using 100 laser shots. The different chain amounts measured by MALDI-ToF MS showed remarkable similarity with HPLC results with some variations, as shown in Table 25. Although Hb F, Hb S and Hb were measurable, it was observed that with the current MALDI-ToF MS instrument the low abundance Hb proportions cannot be measured. The Hb A2 levels and Hb F levels obtained from samples from the sickle thalassaemia patient are listed in Table 24. The spectrum shown in
A sample from known thalassaemia intermedia patient with a HPLC quantification report of the Hb proportions were investigated, as shown in Table 26. It was observed that in this particular instance Hb A2 was measurable but not with confidence. The β and the γ chains show good correlation with the HPLC report. The corresponding spectrum is depicted in
Almost all proteins contain transient or permanent post-translationally modified amino acids such as glycosylated, acetylated, methylated or hydroxylated amino acids. The most common post-translational modification for haemoglobin is glycated Hb whereby the N-terminal valine of the β chain is irreversibly glycated known as the minor Hb A1C fraction. But ESI MS and MALDI-TOF MS studies revealed that glycation occurs in both α and β chains and other glycated proteolytic fragments have been investigated in some reports. The glycation sites of Hb reported by Shapiro et al. show various Val and Lys positions of both the chains as major glycation sites. These post-translational modifications may hinder proteolytic activity.
In this invention, the glycation adducts of patients with different Hb A1C level determined by HPLC method were investigated using the MALDI-ToF MS linear mode. Additionally investigations were carried out to examine if any glycated proteolytic fragments were detectable using on carrier 3 min endoproteinase Glu C digestion in the presence of RapiGest™ at 37° C.
Three whole blood samples having Hb A1C levels of 10.0%, 8.8% and 5.4% and diluted 1:100 with ammonium bicarbonate buffer were screened using the MALDI-TOF MS linear mode. The globin chains and the adducts were resolved with a grid voltage and delay time set to 90% and 350 ns respectively. The resolved m/z values were within 1 standard deviation from the expected masses (listed in Table II), as shown in Table 27.
The peak areas relate to the abundance of an ionic species in MALDI-ToF MS, as such the peak areas for each resolved m/z values were calculated using the Data Explorer software. The percentages for glycated and not glycated globin chains were calculated for individual globin chains and in total by summing all areas of all detected species (100%) and individual species as proportion of the total area, as shown in Table 28. It is evident from
The overlaid MALDI-TOF MS spectra obtained in the linear mode from 5.4% glycated and 10.0% glycated samples show that the peak for the β glycation adduct has a comparatively higher peak height than the α glycation adduct, as shown in
For this invention, the percentages for glycated and not glycated globin chains were calculated for either excluding (Glycation % A) or including the SA adduct area (Glycation % B) to observe the effect of such calculations, interestingly which show that no significant deviation of calculated total glycation percentage occurs if the SA adduct area is left out of the calculation, as shown in
Determination of the presence of glycated peptide peaks and its identification is important for the interpretation of spectra obtained from an on carrier proteolytic digest. To investigate if any glycated globin peaks can be identified, two on Glu C digests were carried out as initial experiments on unpurified EDTA treated blood samples with normal and high glycated Hb proportions (10.0%). The resulting spectra were compared. The same glycated peaks were identified in both the samples but with clearly different signal intensity using the ExPASy FindMod tool, as shown in
In here, two fragments, the glycated and hydroxylated fragment βG8 and the methylated βG3-4 were detected. It was also interesting to observe that the normal counterpart of the peptide fragment, βG8, was not detectable with present experimental conditions, neither for the blood sample with normal nor for the sample with a high Hb glycation level, as shown in
While investigating the peaks it was observed that only one of the glycated peaks, βG8 Gluc-Hydr, whereby the glucose molecule is attached to the β Lys120, has shown a visible difference in the peak obtained from normal and the peak obtained from sample with high glycation. To investigate this finding further, the peak heights, relative intensities, and peak areas of the monoisotopic and most abundant peaks of βG8 Gluc-Hydr were compared with the respective values from the adjacent peak βG4-5. The ratios between the peaks are listed in Table 29 showing an increased ratio for the glycated sample for all three parameters. The appearance of the glycated peptides needs further investigation to confirm its origin, sequence and other relevant mass spectrometric properties.
The MALDI-TOF mass spectra shown in
The effect of trypsin concentration variation for the on carrier digestion of whole human blood in presence of the novel surfactant RapiGest™ was investigated. Although the general effect of shortened digest time on the tryptic fragmentation pattern has been reported, a systematic investigation on trypsin concentration on the fragmentation pattern of the Hb α and β chain is not reported in the literature. In this experiment, the aim was to document the proteolytic fragmentation pattern, optimise on carrier trypsin concentration in relation to the sequence coverage, establish method robustness and check compatibility with automated data analysis.
For this experiment, trypsin stock solution with a trypsin concentration of 1.3 mg/ml (54.5 μM) equalling 5.45 pM/μl was diluted 1:10, 1:20, 1:40, 1:80 and 1:100 fold with 50 mM ammonium bicarbonate buffer, 2 mM CaCl2, pH 8.3. For an on carrier digestion 2 μl of each dilution of trypsin and 2 μl of stock solution without dilution was spotted for each digest on the sample plate and let air dried at room temperature. Three different samples, two blood samples collected from two individuals with normal blood and one blood sample with Hb S, were investigated. For each sample, 3 independent 3 min digests were carried out with the novel method devised in this invention, using the ionic surfactant, on carrier at 37° C. For each digest spot, 10 MALDI-TOF mass spectra were obtained, whereby each spectrum was an accumulation of 5 spectra, each obtained from 100 laser shots. The data were analysed using the Protein Prospector software. It was observed, which adds to the confidence of automated detection of globin chains, that the overall MOWSE score for the detected peptides were high. (MOWSE scores >75 are considered to be significant for protein identification). Although there was a variation in the number of α and βfragments identified within the 10 ppm window, it was constantly higher in trypsin stock solution diluted 1:20 and higher, for normal blood and blood with Hb S variant, as depicted in Table 31.
Interestingly, MALDI-ToF mass spectra obtained for the three samples demonstrate similar fragmentation pattern for each dilution, but they differ in different dilutions. It was observed that a concentration above 1:20 fold stock solution result in loss of bigger tryptic fragments necessary for higher sequence coverage for both chains, which results from a decrease in partial digestion products. To demonstrate this highly significant observation, the clinically important tryptic fragment of βT1 (m/z 952.5098) and partially digested fragments βT2-3(m/z 2228.1669), β1-3 (m/z 3161.6589) were investigated. βT1, sequence positions 1-8, contains Glu at position 6, substitution of which result in Hb S. In the newly established method, in an on carrier digest on normal blood, the βT1, and partially digested fragments βT2-3, βT1-3 fragments are resolved at all time points between 50 s and 3 min.
It was observed that, the m/z values of βT1, βT2-3 and βT1-3 are well resolved with tryptic dilutions from 1:20 to 1:100. Spectra obtained using 1:20 dilution of trypsin is shown in
It was observed that the number of autolytic tryptic fragments decreased as the dilution factor for trypsin increased. With a fixed on carrier trypsin concentration the number of autolytic fragments increased as the dilution factor for sample increased.
Blood from two individuals having normal Hb directly collected in ammonium bicarbonate buffer was subjected to a 3 min on carrier tryptic digests in the presence of the novel surfactant within a few minutes of sample collection and after three weeks. Similar tryptic fragmentation pattern with similar peak intensities, high ion counts, high mass accuracy and excellent mass resolution were obtained from digests performed of these samples at two different time points. Analysis of mass spectra whereby 10 spectra (each an accumulation of 10 individual spectra, each obtained by 100 laser shots) for each digestion were obtained using MALDI-ToF MS reflector mode show a typical fragmentation pattern, as show in
Automated data analysis of an MALDI-TOF mass spectra obtained from a 3 min on carrier digest in the presence of the novel surfactant using the Protein Prospector MS Fit option and the SwissPort.r36 database identified 10 α chain fragments and 9 β chain tryptic fragments within the 10 ppm mass accuracy window, as listed in Table 32. The sequence coverage for the α chain was 70% and the β chain 49% with 10 ppm mass accuracy window.
A blood sample with abnormal peaks identified employing the standard HPLC method was sent for confirmation of diagnosis by DNA analysis to the Clinical Genetic Laboratory at Monash Medical Centre. The sample was obtained from the Monash Medical Centre haematology laboratory for MALDI-ToF MS analysis.
The initial investigation was carried out using the MALDI-ToF MS linear mode. The mass spectrum of obtained from the sample shows the single charged [M+H]+ average m/z value 15127.60 (expected m/z value 15127.37) representative for the α chain with associated error was −0.77 Da, as shown in
An on carrier tryptic digest of the blood sample containing the unidentified β chain variant was obtained with the novel ionic surfactant at 37° C. and a 3 min digest time. 10 MALDI-TOF mass spectra, each an accumulation of 5 spectra whereby each spectrum was obtained by 100 laser shots, were obtained from the digests. Automated data analysis of all the spectra using the Protein Prospector MS Fit programme and the SwissPort.r36 database identified 6-9 α chain tryptic fragments and 5-7 β chain tryptic fragments within the 10 ppm mass accuracy window. The best spectrum with the highest number of identified α and β chain tryptic fragment was identified, baseline corrected, noise filter smoothed and peak deisotoped using the Data Explorer ver 4.0.0.0 software. The deisotoped m/z values were then analysed with two automated data analysis procedures, the FindMod option and the Homology option, the latter using the Protein Prospector programme with molecular mass range set to 15500 to 16000 (β chain mass range), pl 6-7, enzyme to trypsin with maximum missed cleavages to 5, number of amino acid substitution to 1, mass accuracy window to 50 ppm and for the homology mode mass shift to −45.95 Da, −83.75 Da and −121.92 Da respectively. The reproducible occurring unassigned m/z values that were observed for all samples investigated in this study were excluded. The filters were set to exclude to tryptic autolytic fragments and keratin artefact peaks. Only one potential signature peptide was identified with a monoisotopic [M+H]+ m/z value of 1191.6879, as shown in Table 33.
As such, an amino acid substitution that causes a mass shift of −83.0643 Da in the βT4 fragment with a resulting m/z value of 1191.6879 was identified solely by automated data analysis. The substitution identified was Trp (W) to Cyc (C) at position 37 of the β chain as shown in Table 34. No other substitutions were identified at this time point. Simultaneous results reported by DNA analysis using a standard method of the sample aided and confirmed the MALDI-TOF MS identification of the new Hb variant. The reported DNA analysis result was that a mutation in codon 37, G→C (TGG→TGC) had occurred. The presence of normal β chain and normal βT4 tryptic fragment confirms the heterozygous state for the variant.
A blood sample with a HPLC report showing unusual peaks was obtained from the Monash Medical Centre haematology laboratory for MALDI-ToF MS analysis.
Initial investigation carried out using the MALDI-ToF MS linear mode of the sample shows the single charged [M+H]+ average m/z value 15127.65 (expected m/z value 15127.37) representative for the α chain with an associated error of −0.28 Da, a [M+H]+ average m/z value of 15871.12 (expected m/z value 15868.23) representative for the β chain with an associated error of 2.89 Da and an additional poorly separated peak with a m/z value of 15878.98 Da resulting in a mass shift of 10.75 Da.
An on carrier tryptic digest of the blood sample was performed with the novel ionic surfactant at 37° C. and a 3 min digest time. MALDI-TOF mass spectra were obtained using automated data acquisition and 10 collected spectra were analysed using the Protein Prospector MS Fit programme and the SwissPort.r36 database. The best spectrum with the highest number of identified α and β chain tryptic fragments within a 10 ppm mass accuracy window was identified, baseline corrected, noise filter smoothed and peak deisotoped using the Data Explorer software. The deisotoped m/z values were then analysed with two automated data analysis procedures, the FindMod programme and the homology option, the latter using the Protein Prospector software. The criteria were set to a molecular mass range of 15500 to 16000 (β chain mass range), pl 6-7, enzyme to trypsin with maximum missed cleavages to 5, number of amino acid substitution to 1, a mass accuracy window of 50 ppm and for the homology mode mass shift of 5 to 15 Da. Although the obtained mass difference was 10.75 in the linear mode MALDI mass spectrum, mass shifts within a 5 to 15 Da window were explored assuming a poor separation of the β chain peaks resulted in an error in the mass difference between the normal and variant p globin chains. The reproducible, in all spectra of 1:100 dilution of blood occurring unassigned m/z values, possible tryptic autolytic fragments and keratin artefact peaks were not considered using a filter. The automated data analysis identified a signature peptide with a monoisotopic [M+H]+ m/z value of 2072.9705, with 11 possible amino acid substitution for the βT5 tryptic fragment (expected m/z value 2058.9483, received m/z value 2058.9483), as shown in Table 35 corresponding to a 14 Da mass difference.
In the tryptic fragmentation pattern observed for normal blood in this invention the βT5 tryptic region produced a few overlapping fragments. If a mutation occurred, as it is the case with this mutant, resulting in an amino acid substitution causing a 14 Da mass shift in the βT5 fragment, it is expected that this mass shift is also observed in the fragments with missed cleavages. Manual inspection of the spectra confirmed the presence of βT4-5, the βT4-6 (expected m/z values of 3314.6554 and 3541.8187 respectively) and the additional tryptic fragments resulting from the presence of the mutation namely the βTMNO24-5 and the βTMNO24-6 (expected m/z values of 3328.6170 and 3555.8344 respectively), as shown in
The appearance of three signature peptides, as listed in Table 35 confirms the location of the substitution to be in the βT5 tryptic fragment with a mass shift of +14 Da. As such, an amino acid substitution with a list of possible substitution and the location of substitution was identified solely by automated data analysis. From this several possibilities can be excluded. The N→K mutation in not likely, because this would introduce an additional cleavage site and the resulting fragments could not be detected. The D→E mutation can be excluded since the respective fragments could not be detected in the endoproteinase Glu C digests (data not shown). The next step to identify the substitution would have been to perform de novo MS sequencing using CID and PSD analysis. Simultaneous DNA analysis of the sample using standard methods revealed that a mutation at the codon 54, G→C (GTT→CTT) had occurred. The resulting amino acid substitution is Val54 (V)→Leu54 (L) with a mass shift of +14.0157.
The presence of the normal β chain and the normal counterparts of the identified signature peptides βT5NM2, βT4-5NM2 and βT4-6NM2 tryptic fragments, as shown in
Investigations were carried out to optimise conditions suitable for the detection of very low abundance peptides in a complex mixture of high and low abundance peptides derived from on carrier digests of protein mixtures. Blood contains a complex mixture of Hbs with a high abundance of Hb A (α2β2). The Hb A2, a minor component of adult blood has two δ chains with two α chains (α2δ2), and consists of only 2-3% of the total Hb content, where as the Hb F (α2γ2), another minor component is present in adults only in trace amounts (less than 1%). The δ chain percentage equals the Hb A2 percentage. The level of ζ chain in normal newborns averages 0.19% although it varies considerably with ethnicity. Thus, a proteolytic digest of whole blood would yield a very complex mixture of their peptides derived from all the Hb chains with various abundances making the identification of proteolytic peptide fragments very difficult and challenging.
Normal blood with adult Hb diluted 1:100 was incubated with the novel detergent RapiGest™ for 5 minutes and diluted 1:500, 1:1000, 1:5000, 1:10000, 1:50000 and 1:100000 with ammonium bicarbonate buffer followed by the newly developed method for on carrier 3 min proteolytic digestion at 37° C. for each dilution. For each dilution 5 different spectra were accumulated, each with an accumulation of 10 spectra whereby each spectrum was an accumulation of 100 laser shots. All the spectra were thoroughly analysed by visual inspection and automated protein identification using the Protein Prospector software. The appearance and disappearance of certain peaks were monitored for all the dilutions. The signal strength was determined by calculating the signal to noise ratio using the Data Explorer software. The on carrier 3 min tryptic digest at 37° C. in the presence of the novel surfactant RapiGest™ produced strong signals for the αT4, αT2-3 and the βT4 proteolytic fragments (m/z values 1529.7342, 974.5418 and 1274.7255). Initially, these three peaks were monitored for their appearances for all the dilutions. All three the peaks were detectable with confidence for dilutions as high as 100000, although the signal strength gradually decreased, as shown in Table 37,
In this invention the 69-17 fragment was monitored to monitor the effect of the dilution factor on a low abundance Hb A2 fragment. It was interesting to observe, that the signal strength for the peak gradually increased as the dilution factor was increased reaching its highest strength in the 1:100000 dilution, as shown in Table 37 and
Three different dilutions of blood samples obtained from three patients having α gene deletions --/αα(-α3.7/-α3.7, -α3.7/--SEA) and one normal Hb from blood of a healthy individual, 1:10, 1:100 and 1:1000 with ammonium bicarbonate buffer, were investigated. The on carrier trypsin digestion of these samples was performed with the presence of the ionic surfactant RapiGest™ SF at 37° C. and a 3 min digest time. For each sample, 10 accumulations, each for 5 and 50 spectra, were obtained. Each spectrum was obtained by 100 laser shots (laser intensity set to 2400), and accumulated using selection criteria of a minimum resolution of 10000, a minimum signal intensity of 1000 and a maximum signal intensity of 64000 for the base peak, βT4 (1274-1275). All spectra were analysed using the ProteinProspector software, and for the automated detection of Hb ζ chain, the pre-processing filter was set to a mass accuracy of 400 ppm and the post-processing filter was set to a final mass accuracy of 250 ppm, the mass range to 5000-16500 Da, and the pl to 6.5-9. The results obtained for the two α gene deletion samples of three different dilutions were compared against the normal.
Analysis of the obtained spectra of the samples, as shown in Table 38, demonstrate that with the condition applied in this invention, the detection of the following ζ tryptic fragments were possible, with increasing mass, ζT8 (m/z 928.5642), ζT3 (m/z 1048.5859), ζT5 (m/z 1070.5993), ζT9 (m/z 1075.5629), ζT6 (m/z 1885.9343) and ζT14 (m/z 1308.7409). Since the αT11 and the ζT11 both have the same amino acid composition, and as such posses the same m/z value, 818.4406, it was not considered as a diagnostic fragment, although it was detected.
and
Some peptide fragments derived from the minor Hb fractions, the γ and chains, were also detected. The detected δ chain fragments, derived from minor Hb component A2, with increasing mass, were δT3 (m/z 1256.6593), δT14 (m/z 1441.6780), δT13-14 (m/z 1887.9058), δT2-3 (m/z 2197.1723) and δT114-15 (m/z 3018.5618). The δT15 (m/z 1149.7961.) which has an identical m/z value as βT14, the δT4 having a identical m/z value with βT4/εT4/γT4 (m/z 1274.7255), δT9 (m/z 1669.891) with βT9, δT14-15 with a m/z value similar to βT14-15 (1449.7961 and 1449.008 respectively), δT9 identical with βT9 (m/z 1669.8907) and δT8-9 identical with βT8-9 (m/z 1797.9857) were also detected, as shown in Table 39.
The detected y chain fragments identified unambiguously, derived from Hb component F, present in trace amount in adults, with increasing mass, were the γT1 (m/z 1093.4624 with MetINI) and the γT12 (m/z 3124.7193). The γT111 fragment (m/z 1098.5578) is identical to the εT11, the γT4 having an identical m/z value with βT4/εT4 (m/z 1274.7255), the εT2-3 with a m/z value similar to the δT5-6 (2274.1724 and 2272.0954 respectively) were also detected, as shown in Table 40.
After automated analysis and detection of peaks, all the spectra were manually inspected to confirm the presence of the respective peak. The comparison of the 50 accumulated spectra with 5 accumulated spectra show that an increased number of ζ chain fragments were identified with greater dilution of the sample, in particular the 1:1000 dilution, and that the ζT3 and the ζT5 were identified in all three samples with α thalassaemia in all dilutions when 50 spectra were accumulated, as shown in Tables 41 and 42. The mass accuracy of the identified ζ chain fragments was low, which is expected because of the extremely low abundance of the ζ chain fragment ions. The presence of the ζT3 and the ζT5 in all three dilutions when 50 spectra were accumulated are shown in
Finally it is to be understood that various other modifications and/or alterations may be made without departing from the spirit of the present invention as outlined herein.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2004902922 | May 2004 | AU | national |
| 2004903001 | Jun 2004 | AU | national |
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/AU05/00755 | 5/27/2005 | WO | 00 | 8/22/2007 |
