Novel nucleic acid purification chemistry

Abstract

The present invention relates generally to the field of nucleic acid isolation on silica solid supports. Specifically, a novel silica solid support nucleic acid binding buffer chemistry is disclosed herein, based on the use of small quaternary organic compounds, such as tetramethylammonium chloride (TMAC), under acidic conditions. This novel nucleic acid purification chemistry has the potential to purify not only RNA but also DNA and be implemented in a wide range of commercial kits, from spin columns to integrated lab-on-a-chip (LOC) devices, such as disposable cartridges using solid-phase extraction technology. Furthermore, the method of the present invention can be performed using a relatively small volume of binding buffer, potentially enabling increased sample input volumes in such integrated or enclosed molecular diagnostic devices and increasing the chances of detecting trace amounts of nucleic acid targets in liquid biopsy samples, such as plasma or urine.

Description

[Technical Field]

[0001] The present invention relates generally to the field of nucleic acid isolation on silica solid supports. Specifically, a novel silica solid support nucleic acid binding buffer chemistry is disclosed herein, based on the use of small quaternary organic compounds, such as tetramethylammonium chloride (TMAC), under acidic conditions. This novel nucleic acid purification chemistry has the potential to purify not only RNA but also DNA and be implemented in a wide variety of commercial kits, from spin columns to integrated lab-on-a-chip (LOC) devices, such as disposable cartridges using solid-phase extraction technology. Furthermore, the method of the present invention can be performed using a relatively small volume of binding buffer, potentially enabling increased sample input volumes in such integrated or enclosed molecular diagnostic devices and increasing the chances of detecting minute amounts of nucleic acid targets in liquid biopsy samples, such as plasma or urine. [Background technology]

[0002] In a patent application filed in 1990, U.S. Patent No. 5,629,669, Boom et al. described a nucleic acid purification technique based on adsorption to a general-purpose solid support. Boom's extraction uses a large amount of chaotropic salt, with or without alcohol, to mediate the binding of nucleic acids to silica. Due to its high performance, enabling nucleic acid extraction yields of over 50% from biological samples, it quickly became the gold standard in nucleic acid isolation and continues to be widely used to this day in many commercially available extraction kits and integrated molecular diagnostic devices. For example, Boom's protocol, or some variation thereof, forms the basis of the DNA extraction principle used in QIAGEN's QIAamp Circulating Nucleic Acid kit or Biocartis NV's integrated cartridges, such as IdyllactRAS.

[0003] Boom's protocol requires a large amount of binding buffer relative to the amount of biological sample, due to the need for a large amount of chaotropic salt followed by additional alcohol. Therefore, given the continued rise of ever-smaller, handheld, fully integrated lab-on-a-chip (LOC) molecular testing devices, which often lack sufficient buffer reservoirs due to their desire to maximize sample volume, there is a need to find an efficient alternative to Boom's protocol. Another reason is that chaotropic salts are expensive and have strong PCR-inhibiting properties, which can pose numerous challenges for the final product manufacturing line. For all of the above, there is currently a need for efficient chaotropic-free nucleic acid purification chemistries that allow for increased sample input, especially in integrated systems used in the field of liquid biopsy, where sample size is increasingly important for detecting minute quantities per milliliter of plasma cell-free (cf) DNA target.

[0004] To date, several attempts have been made to develop novel chaotropic-free silica-based nucleic acid purification chemistries. Notable examples include:

[0005] A method describing the purification of viral RNA based on acidic conditions and the use of kosmotropic salts was published by Hourfar et al. in 2005. The publication was titled "High-Throughput Purification of Viral RNA Based on Novel Aqueous Chemistry for Nucleic Acid Isolation." This method is specific to RNA and is not suitable for purifying DNA from plasma.

[0006] A similar method involving the isolation of total RNA from E. coli by using kosmotropic Hofmeister salt was published by Lee et al. in 2008. Additionally, Lee et al. hold a patent, U.S. Patent No. 5,623,999, entitled "Method of purifying RNA using kosmotropic salt." Both these publications and patents describe and focus on RNA-selective purification chemistry based on the use of acidic conditions and kosmotropic salts, but provide no teaching on how to apply this chemistry to DNA.

[0007] Johns Hopkins University has a patent application, U.S. Patent No. 5,629,999, entitled "Chaotrope- and volatile-free method for purifying nucleic acids from plasma." The method described therein is very similar to that of Lee et al. and involves the use of acidic conditions and kosmotropic salts to mediate binding of RNA to silica.

[0008] MiDiagnostics owns a very similar patent application, U.S. Patent No. 6,299,499, entitled "System and method for purifying and amplifying nucleic acids." This patent application describes a nucleic acid purification chemistry that uses acidic conditions and kosmotropic salts to mediate binding of viral nucleic acids to silica. However, this method provides no evidence that it is at least as efficient as Boom's protocol, nor that it is applicable to purifying cfDNA present in plasma.

[0009] Despite its widespread use in molecular diagnostics, the interaction of nucleic acids with silica is still poorly understood. In fact, not much has changed since Boom et al. described the first silica-based nucleic acid purification technique in 1990. To date, there have been very limited studies attempting to elucidate the fundamental mechanics of DNA / RNA adsorption to silica.

[0010] (2003) is one of the few to attempt to demystify the underlying mechanics of Boom's extraction technique. They describe three effects that are thought to be the main drivers of nucleic acid adsorption to silica: Screening of intermolecular electrostatic forces, Dehydration of DNA and silica surfaces, the formation of intermolecular hydrogen bonds at the nucleic acid-silica contact layer (described as the least dominant trigger); Examples include:

[0011] The above three factors can be controlled by adding various salts to nucleic acid solutions in the presence of silica solid supports. Hofmeister classified salts based on their ability to affect the structure of macromolecules, primarily proteins, in these aqueous solutions. According to this classification, chaotropic salts were initially described as structure-breaking agents, as they increase protein solubility (the so-called "salting-in" process). In contrast, kosmotropic salts were described as structure-forming agents, as they decrease protein solubility (the so-called "salting-out" process). In the context of silica-based nucleic acid isolation, chaotropic salts are a natural choice due to their ability to affect water structure and cause a dehydration effect. From this perspective, chaotropic ions are described by Hofmeister as large, monovalent ions with low charge density that interact less with water than with water itself. Chaotropic ions are believed to have minimal interference with the hydrogen bonding of surrounding water. For example, the original nucleic acid purification chemistry described by Boom et al. in 1990 used high concentrations of guanidinium thiocyanate due to its strong chaotropic properties, its lytic properties, and its potential to inactivate ribonucleases. In contrast, kosmotropic ions are described as small or multivalent ions that have a high charge density and are therefore capable of breaking water-water hydrogen bonds.

[0012] Although guanidinium cations and thiocyanate anions are not expected to have large hydration shells, their excessive concentrations (3 M–5 M) used in Boom's protocol are thought to compensate for this. It is hypothesized that this high salt concentration can sufficiently reduce the concentration of free water, thereby causing dehydration of the nucleic acid and the silica membrane. Furthermore, the abundance of guanidinium cations is thought to shield the electrostatic forces between the negatively charged phosphate backbone in the nucleic acid and the negatively charged silanol groups on the silica surface. Both of these effects may be hypothesized to promote hydrophobic interactions between the bases and the siloxane bridges, thereby enabling nucleic acid adsorption to the silica membrane. Subsequent adaptations of Boom's original protocol include adding alcohol to the binding buffer to further reduce the concentration of free water and enhance this dehydrating effect.

[0013] Following this modification, the silica-bound nucleic acids are then washed with strong alcohol (often 70%-90% ethanol). This washing procedure ensures the removal of residual non-nucleic acid compounds from the biological sample or binding buffer. Finally, the nucleic acids are eluted using a low ionic strength solution at a neutral or slightly basic pH. This elution mechanism allows for direct compatibility with downstream applications such as PCR and NGS.

[0014] As explained above, Hourfar et al. were the first to describe an alternative silica-based approach for purifying RNA from biological samples. Subsequent publications and / or patents from Samsung Electronics, Johns Hopkins University, and MiDiagnostics are based on the same chemistry, using acidic conditions and kosmotropic salts to mediate the binding of RNA (and, to a much lesser extent, DNA) to silica. A possible explanation for why this chemistry works with RNA may be based on the following: silanol groups on silica surfaces have pKa values in the range of 4–8. Lowering the pH of the binding buffer / sample mixture below these values favors the protonation of the weakly acidic silanol groups, thereby eliminating their strong negative charge. As a result, electrostatic repulsion by the negatively charged phosphate backbone of nucleic acids is significantly reduced, or even completely eliminated.

[0015] Furthermore, nucleic acids and silica membranes can be dehydrated using minimal amounts of kosmotropic salts (i.e., (NH4)2SO4) to significantly reduce the amount of free water. As explained above, strong kosmotropic ions have large hydration shells and can therefore capture significant amounts of free water. In that regard, it can be hypothesized that only limited amounts of kosmotropic salts (400 mM to 1000 mM, depending on the particular salt) are required to produce an effect similar to, for example, 5 M guanidine thiocyanate. These effects can be used to explain the binding of flexible RNA to silica and, to a much lesser extent, the binding of double-stranded, and therefore more rigid, DNA to silica.

[0016] The bound nucleic acids are then washed with a high percentage of alcohol as in the Boom protocol, although a variant has been described in which the washes are completely alcohol-free. In these variants, the washes are performed with a buffer similar to the binding buffer or a simplified version thereof, i.e., an acidic solution (pH 4-7) without or containing a limited amount of kosmotropic salts. It can be assumed that these washing measures are primarily based on an attempt to prevent elution of nucleic acids by eliminating electrostatic repulsion through protonation of silanol groups. The elution mechanism is then similar to that in the Boom protocol.

[0017] While this approach has proven very successful for purifying RNA (and is even sometimes described as RNA-selective), it should be noted that purifying double-stranded DNA (dsDNA) remains much more challenging. Extraction yields of dsDNA with these chemistries have been shown to be 10- to 100-fold lower, which is clearly insufficient for extracting DNA from plasma.

[0018] Herein, the disadvantages of nucleic acid extraction methods free of strong chaotropic agents have been addressed by the successful use of salts consisting of small quaternary organic compound cations and very weakly chaotropic, highly soluble anions. Quaternary compounds are cations consisting of a central positively charged atom with four uncharged substituents, mostly alkyl and aryl groups. These cations are permanently charged regardless of the pH of their solution. These cations are often described as inert cations. In particular, salts such as tetramethylammonium (TMA) + ) cation and weakly chaotropic chloride (Cl - ) anion or bromide (Br - We have observed and demonstrated that salts consisting of a combination of dsDNA and a hydroxyl group containing hydroxyl group (H) anions produce unique conditions for isolating dsDNA on silica solid supports at acidic pH with efficiency comparable to chaotropic agent-based protocols such as Boom's protocol.

[0019] To the best of our knowledge, salts consisting of small quaternary organic cations and weakly chaotropic anions have not been used in solid-support-based nucleic acid extraction from biological samples. In 1977, J.M. Orosz and J.G. Wetmur (Non-Patent Document 2) used similar salts to study the melting and renaturation properties of DNA and dsRNA, but their study did not consider the option of using these salts in solid-phase nucleic acid extraction. Furthermore, U.S. Patent No. 5,629,297 describes the use of sodium salts in combination with quaternary ammonium salts in a method for enriching nucleic acids containing single-stranded poly(A) stretches (i.e., primarily messenger RNA) while removing undesired nucleic acids, such as rRNA, on a solid support coated with immobilized oligo-dT capture probes. The teachings of U.S. Patent No. 5,629,297 imply strict and selective specificity for poly(A) nucleic acids and do not appear to be suitable for isolating any other types of nucleic acids from samples such as liquid biopsies. Next, U.S. Patent No. 6,263,666 discloses a hybridization solution containing tetramethylammonium chloride ((CH3)4NCl) and a cationic detergent that immobilizes synthetic oligonucleotides on a solid surface such as polystyrene. Importantly, however, U.S. Patent No. 6,263,666 does not teach the purification of native nucleic acids on a solid support from the context of complex biological samples, particularly from liquid biopsy samples such as plasma. In conclusion, none of the above disclosures teach or suggest the use of small quaternary organic salts for nucleic acid isolation as a general alternative to the Boom protocol.

[0020] The approach presented herein offers several advantages over Boom's protocol. First, Boom's protocol was originally designed to isolate long genomic and plasmid DNA / RNA. The highly fragmented nature of genomic material typically present in plasma and FFPE samples poses challenges to its use in molecular diagnostics. Short nucleic acid fragments naturally have far fewer hydrophobic binding sites, limiting their binding efficiency to silica under the high concentrations of chaotropic agents used in Boom's protocol. Contrary to this limitation, the approach presented herein, based on salts composed of quaternary cations and mild chaotropic agents, was able to isolate both short and longer dsDNA fragments. Furthermore, it was observed that the extraction efficiency of short DNA (i.e., with lengths ranging from 10 bp to 300 bp) can decrease with increasing pH, while the extraction efficiency of high-molecular-weight DNA can increase with increasing pH. Therefore, by varying the pH value, the approach disclosed herein further opens up the opportunity to fine-tune the extraction efficiency depending on the desired DNA target length.

[0021] A second advantage of the method disclosed herein is that the chemical composition of the binding buffer does not cause protein aggregation, which in principle allows plasma samples to be processed without a protein digestion step. The mixture of plasma and binding buffer allows for a smooth flow rate, facilitating its use in microfluidic devices. Therefore, although the method disclosed herein can be used to omit the protein digestion step for small-volume or dilute samples, for some older plasma samples with a volume greater than 400 μL, including the protein digestion step as an optional step may still be useful to increase the final extraction yield.

[0022] Second, as mentioned above, LOC devices or disposable cartridges often lack sufficient storage for the relatively large volumes of binding buffer used in Boom's protocol. Furthermore, chaotropic salts are expensive, have strong PCR-inhibiting properties, and can cause numerous problems in manufacturing lines, for example, due to crystallization. In contrast, the methods and binding buffers presented herein do not contain strong chaotropic agents, are inexpensive, and significantly reduce the volume of binding buffer required per sample, thereby enabling increased sample input in fully integrated molecular diagnostic devices. This is a major advantage, for example, when processing liquid biopsies obtained from cancer patients, where the amount of tumor-derived mutant DNA copies per milliliter of plasma is very low and difficult to detect. Finally, the methods presented herein are comprehensive, meaning they enable the efficient purification of both short and long dsDNA, as well as ssDNA and, in some cases, RNA, from a variety of biological samples. [Prior art documents] [Patent documents]

[0023] [Patent Document 1] European Patent No. 389063 [Patent Document 2] U.S. Patent No. 7,923,551 [Patent Document 3] International Publication No. 2016 / 073824 [Patent Document 4] International Publication No. 2018 / 156906 [Patent Document 5] International Publication No. 2015 / 165859 [Patent Document 6] International Publication No. 1995 / 015970 [Non-patent literature]

[0024] [Non-Patent Document 1] Melzak et al. (1996) [Non-patent document 2] Biopolymers, vol. 16, 1183-1977 Summary of the Invention

[0025] The chemistry of silica-based purification of nucleic acids has not changed significantly since Boom et al. published the original method in 1990. In many cases, this is due to the adequate performance of Boom's extraction technique, which results in nucleic acid extraction yields of over 50% from biological samples. Boom's protocol mediates the binding of nucleic acids to silica by using significant amounts of chaotropic salts and alcohol. However, in light of the recent rise of miniaturized, integrated lab-on-a-chip (LOC) devices, it seemed necessary to minimize the buffer volumes contained therein and maximize the sample volume input they can accommodate. Thus, there is a clear need for novel nucleic acid purification chemistries that are free of strong chaotropic agents, are economical, and allow for increased sample input in integrated systems.

[0026] Presented herein is a novel purification chemistry that mediates the binding of nucleic acids to silica using only acidic conditions and relatively small amounts of salt composed of quaternary ammonium compounds. The disclosed method significantly reduces the volume of binding buffer relative to the sample volume, allowing for increased sample input, which is highly beneficial in fully integrated molecular diagnostic devices. Furthermore, the disclosed isolation method results in reasonable nucleic acid yields that rival the performance of strong chaotropic agent-based protocols such as Boom's protocol.

[0027] For a more complete understanding of the concepts presented herein, reference is made to the following detailed description taken in conjunction with the accompanying drawings. [Brief explanation of the drawings]

[0028] [Figure 1]FIG. 1 shows the extraction efficiency of dsDNA in various chaotropic and kosmotropic salt binding buffers at neutral pH. [Figure 2] FIG. 1 shows the extraction efficiency of dsDNA in various chaotropic and kosmotropic salt binding buffers at acidic pH. [Figure 3] FIG. 1 shows a comparison between Boom's extraction binding buffer and a chloride-based buffer with or without a quaternary ammonium compound. [Figure 4] FIG. 1 shows a comparison of different DNA extraction chemistries at different pH ranges. [Figure 5] FIG. 1 shows a comparison between TMAS and TMAC. [Figure 6] FIG. 1 shows the performance of various TMA-containing salts. [Figure 7] FIG. 1 shows the performance of TMAC at various binding concentrations. [Figure 8] FIG. 1 shows the performance of TMAC at various pH values. [Figure 9] FIG. 1 shows the performance of TMAC on different plasma batches with or without CTAB at different concentrations. [Figure 10] FIG. 1 shows the performance of various TMAC buffers. [Figure 11] FIG. 1 shows an investigation of various elution conditions. [Figure 12] FIG. 1 shows the potential benefits of including proteinase pre-digestion. [Figure 13] FIG. 1 shows the performance of the method in a sealed integrated cartridge. [Figure 14] FIG. 1 shows a comparison between the extraction of Boom and TMAC+CTAB extraction chemistries on different plasma batches in a sealed integrated cartridge. [Figure 15] FIG. 1 shows a comparison between the extraction of Boom and TMAC+CTAB extraction chemistries on different plasma batches in a sealed integrated cartridge. DETAILED DESCRIPTION OF THE INVENTION

[0029] The present disclosure generally relates to a method for preparing a biological sample, optionally a liquid biopsy sample, and a silica solid support at a pH value between 3 and 6, and small quaternary organic compounds, defined as quaternary compounds consisting of a central positively charged atom bearing four organic substituents R1-R4, where the number of carbon atoms in each of the organic substituents R1-R4 does not exceed two; a bromide anion or a chloride anion; The present invention relates to a method for extracting nucleic acid, comprising contacting a nucleic acid with a base in the presence of a salt consisting of:

[0030] In other words, a novel binding buffer chemistry based on acidic conditions and the use of minimal salts with small quaternary organic compounds is disclosed herein, providing a comprehensive, chaotropic-agent-free nucleic acid purification protocol that enables the efficient isolation of not only RNA but also DNA. As used herein, the term "quaternary compound" should be used interchangeably with the term "quaternary organic compound," which should be understood as a chemical compound defined as a cation or having an ion that is a cation consisting of a central positively charged atom bearing four organic substituents (i.e., alkyl and / or aryl groups and the balance hydrogen atoms), further designated organic substituents R1-R4. As used herein, the term "small quaternary organic compound" should be understood as a quaternary organic compound in which the number of carbon atoms in each of the four organic substituents R1-R4 does not exceed two carbon atoms. Considering solubility, the preferred organic substituent is a single-carbon group, i.e., a methyl group. Working with small quaternary organic compounds is easier and more preferred because the more often the four organic substituents R1-R4 consist of methyl groups, the higher the solubility is considered to be. Notwithstanding the above, it is also believed that organic substituents containing two carbon atoms in one or more of the four organic substituents R1-R4 are sufficiently soluble and quite suitable for carrying out the methods disclosed herein.

[0031] The most well-known quaternary compounds are quaternary ammonium cations (R4N) with a central nitrogen atom. + ) are quaternary ammonium salts, which are salts containing quaternary ammonium salts. Thus, in one embodiment, a method is provided in which the positively charged atom of the small quaternary organic compound is nitrogen. Other possible examples and reasonably workable embodiments include quaternary phosphonium salts (R4P + ), quaternary arsonium salts such as arsenobetaine (R4As + ), as well as some arsenic-containing superconductors. + ) and bismuth salts (R4Bi + ) are also described as being present and may possibly function in certain embodiments of the methods presented herein.

[0032] As shown in the following examples, the anion of the salt used in the method of the present invention also affects the final nucleic acid extraction yield. Unlike the known RNA-specific methods described above, strongly kosmotropic anions do not appear to be suitable for DNA extraction. Instead, it has been found that very weakly chaotropic ions such as bromide or the even weaker / marginally kosmotropic chloride ion generally produce the best results, with the latter being slightly preferred in most experimental configurations. Thus, in the following embodiment, a method is provided in which the anion is chloride.

[0033] It is hypothesized that the differences highlighted above between RNA and DNA binding to silica may be due to the at least partially single-stranded nature of RNA. That is, RNA binding to silica may be easier than dsDNA binding, possibly due to the increased rotational mobility of bases in single-stranded nucleic acids, thereby increasing the amount of available hydrophobic binding sites for them. Conversely, for double-stranded DNA, significant changes in its helical structure may be required to facilitate hydrophobic interactions between the bases and the siloxane (Si-O-Si) bridges of the silica membrane.

[0034] The DNA double helix is mainly hydrogen bonding between the base and the aqueous environment, electrostatic shielding of the negatively charged phosphate backbone, base-stacking interactions between adjacent bases, is thought to be stabilized by

[0035] The latter has been explained as the most dominant contributor to double helix stability: destabilization of the double helix is required to allow efficient binding to the silica membrane, and it has been hypothesized that the type and amount of ions present play a major role in defining the helical conformation of double-stranded DNA.

[0036] Based on our experience, acidic conditions and kosmotropic salts do not appear to promote dsDNA binding to silica. Using limited amounts of kosmotropic salts may alter the conformation of the double helix, possibly further reducing the affinity of dsDNA for silica membranes. For this reason, we hypothesize that kosmotropic agent-based methods known prior to this approach are RNA-selective. Small cations with high charge density could theoretically fit between the minor and major grooves of the helix structure, whereas strongly kosmotropic anions strongly dehydrate the double helix, possibly causing it to conformationally change to rigid A-DNA, thereby reducing the availability of bases for binding on the silica solid support. Following this reasoning, we hypothesized that to counteract this affinity-reducing effect and promote dsDNA binding to silica membranes, the stabilizing effect of cations should be eliminated and the dehydrating effect of kosmotropic anions should be reduced. It has been found that this effect can be achieved by using salts composed of quaternary ammonium compounds and weakly chaotropic / weakly kosmotropic anions, such as chloride, which have been described as being on the border between chaotropic and kosmotropic behavior. Quaternary ammonium compounds are cations consisting of a central positively charged nitrogen atom bearing four uncharged substituents, mostly alkyl and aryl groups. These cations are permanently charged regardless of the pH of the solution. These cations are often described as inert cations.

[0037] Based on our theoretical model, we hypothesize that the success of this approach using small quaternary organic compounds such as TMAC can be attributed, at least in part, to the inertness and overall bulk of the quaternary ammonium cations preventing electrostatic shielding of the negatively charged phosphate backbone of dsDNA. +It is also possible that the methyl groups of α- and β-dsDNA may cause steric hindrance, preventing its binding to the minor or major groove of the helix. While the use of salts composed of such inert cations prevents conformational changes that adversely affect the affinity of dsDNA for silica, it is conceivable that the weakly chaotropic anions may still dehydrate the silica support sufficiently for the observed efficient binding of the dsDNA double helix.

[0038] In keeping with the above, in the next embodiment, a method is provided in which the small quaternary organic compound is tetramethylammonium chloride, further referred to as TMAC. In a further embodiment, as evidenced in the examples below, the concentration of the small quaternary organic compound, exemplified as TMAC concentration under silica binding conditions, is comprised between 0.1 M and 2 M, possibly between 0.5 M and 1.8 M, possibly between 0.8 M and 1.6 M, possibly between 1 M and 1.4 M, or can be about 1.2 M.

[0039] One advantage of the disclosed methods is their potential maximization of sample input volume relative to the required amount of binding buffer, depending on the desired application. This feature is particularly advantageous for integrated devices such as sealed fluidic cartridges with a defined and limited internal volume. This feature directly depends on the solubility of the binding buffer components. The preferred small quaternary organic compound TMAC has excellent solubility of over 1000 g / L, corresponding to a stable 9 M TMAC solution at room temperature. Sodium acetate, an exemplary buffer compound ensuring acidic pH conditions, also has high solubility in water, equivalent to 5.6 M. Thus, for example, in one embodiment of the disclosed method, when 1.2 M TMAC and 0.2 M acetate are used in the conditions for binding nucleic acids to silica (i.e., the conditions for contacting the sample and binding buffer with the silica solid support), a sample-to-buffer ratio of 6 / 1 can be achieved in the binding conditions. Such an exemplary binding buffer contains 8.4 M TMAC and 1.4 M acetate, both of which are soluble at room temperature. Thus, in possible embodiments, the sample / buffer ratio can range from 6 / 1 to 1 / 6 overall, depending on suitability for the application.

[0040] The uniqueness of the disclosed approach lies in the use of salts composed of small quaternary organic compounds and weakly chaotropic anions (i.e., chloride ions, and to some extent bromide ions) to mediate the binding of nucleic acids (not only RNA but also DNA, especially dsDNA) to silica membranes under acidic conditions. As used herein, the term "acidic conditions" is understood to refer to conditions in aqueous solution where the pH, as estimated on a standard pH base 10 logarithmic scale of hydrogen ion molarity (measured in moles per liter), is at least below a value of 7. Accordingly, in further embodiments, methods are provided in which the pH is comprised between 4 and 5.8, between 4.2 and 5.6, between 4.4 and 5.4, between 4.6 and 5.2, and likely about 5. Given the specific salt compositions described above in these pH ranges and in the presence of silica solid-bound supports, it is believed that a comprehensive nucleic acid purification technique is provided that is compatible not only with RNA but, importantly, with DNA.

[0041] Following binding to silica, the nucleic acids can then be washed and eluted by standard silica washing and elution methods known in the art.

[0042] For example, washing of bound nucleic acids can be performed in a manner similar to the methodology used in the original Boom protocol. That is, strong alcohol, often 90% ethanol, can be used. As discussed above, some previously known methods described washing procedures based on acidic solutions without kosmotropic salts or containing minimal amounts of kosmotropic salts. Based on our observations, such approaches are believed to be compatible only with RNA applications. Rehydration of dsDNA during washing is believed to cause stabilization of the double helix through hydrogen bond formation between bases, resulting in its premature release from the silica solid support.

[0043] The elution mechanism is then largely similar to that known from previously known methods. For example, a solution of low ionic strength at a neutral or slightly basic pH is used. This can be either water or a standard PCR buffer. It has also been observed that the pH and amount of divalent cations in the elution buffer can significantly affect elution efficiency. This is likely because charge repulsion between negatively charged silanol groups and the negatively charged phosphate backbone likely plays an important role during elution. Therefore, deprotonation of silanol groups (strengthening of negative charge) by increasing the pH of the elution buffer leads to improved elution efficiency. The complete absence of small cations and / or divalent cations also promotes elution efficiency, likely due to the lack of electrostatic shielding. These mechanisms are generally known in the art, and therefore, selecting a washing and elution strategy appropriate for a protocol is not a major challenge for those skilled in the art and will not be discussed further herein.

[0044] In alternative further embodiments, since abandoning the strong chaotropic chemistry of the Boom protocol, at least for some types of biological samples, can pose some challenges, several additions to the methods disclosed herein can be employed.

[0045] Specifically, focusing on Boom's extraction-based protocol, chaotropic salts are: (i) preventing other biomolecules (e.g., proteins, lipoproteins) from depositing on the silica solid support; (ii) inhibiting the activity of nucleases; (iii) releasing DNA from histone proteins to enhance its interaction with silica; This makes it possible.

[0046] As known by those skilled in the art, the same effect can be achieved by introducing a protease to perform a protein digestion step, which can be advantageous in particularly difficult (e.g., old) samples. Thus, in another embodiment, the method is preceded by a protease treatment, for example with proteinase K.

[0047] In another embodiment, a method is provided in which the biological sample is a liquid biopsy sample. As used herein, the terms "liquid biopsy" or "liquid biopsy sample" are understood to refer to any non-tissue specimen, particularly a bodily fluid sample, obtained from a subject. Sources of liquid biopsies include, but are not limited to, blood, plasma, serum, urine, cerebrospinal fluid (CSF), amniotic fluid, and other bodily fluids, such as saliva, sweat, tears, milk, semen, feces, pleural fluid, peritoneal fluid, or lavage fluid. Analyzing nucleic acids in liquid biopsy samples can minimize the need for expensive, often invasive, and painful tissue and / or tumor biopsies that enable dynamic disease or other physiological condition monitoring. For example, in cancer patients, cell-free tumor DNA or RNA extracted from liquid biopsies can potentially be used to detect mutations, translocations, or copy number changes, as well as the expression of specific cancer markers.

[0048] Blood (as well as plasma, serum, or whole blood) is the most commonly described fluid used for the analysis of liquid biopsy samples in humans. In cancer patients, blood is a source of circulating tumor cells (CTCs) and cell-free DNA (cfDNA) and cell-free RNA (cfRNA), including circulating tumor DNA (ctDNA) and circulating tumor RNA (ctRNA), respectively, released from tumor tissue, which can be used to detect mutations present in the patient's tumor. Notably, however, ctDNA comprises only a small fraction of the cfDNA present in blood, highlighting the importance of maximizing sample volume for nucleic acid analysis to detect rare mutations. Furthermore, cfDNA is always of low quality and is fragmented to approximately the size of a nucleosome (140 bp). Therefore, for certain cancer types, including kidney, prostate, and upper and lower tract urothelial carcinomas, urine may be a more abundant source of tumor-derived material, and therefore alternative liquid biopsy approaches are used. Urine also has other unique benefits, such as ease of acquisition (no need for skilled medical staff), no patient discomfort (improving patient compliance), and may contain fewer contaminating proteins than blood. However, urine is still a very dilute material, so its use in diagnostic approaches, particularly in PoC devices, would also benefit from maximizing sample input volumes. In view of the existing need for nucleic acid extraction chemistries that allow for maximizing blood or urine sample volume input, particularly within integrated PoC devices such as fluidic cartridges, a method is provided in another exemplary embodiment, in which the liquid biopsy sample is selected from plasma, serum, whole blood, or urine, as the method is well suited for this purpose.

[0049] In a related embodiment, methods are provided in which the nucleic acid is DNA, which is more stable than RNA despite being relatively dilute in a liquid biopsy sample, and which can be isolated using the methods disclosed herein with efficiency similar to that of the Boom extraction-based protocol.

[0050] In further embodiments, the DNA may be cell-free DNA (cfDNA) or circulating tumor DNA (ctDNA), which is typically in the form of fragmented and / or double-stranded DNA.

[0051] Notably, with certain whole blood or aged plasma or serum samples, the novel binding chemistry presented herein (e.g., 1 M TMAC + 0.2 M acetate at pH 5) can sometimes be problematic due to excessive protein precipitation and / or blocking of the silica solid support, which can result in reduced extraction yields. Depending on the sample type, the addition of an appropriate detergent can address this issue. Thus, in another alternative embodiment, a method is provided in which contacting is performed in the presence of a detergent. As used herein, the term "detergent" should be broadly interpreted as relating to a chemical compound or mixture with surface-active properties. As used herein, the term "detergent" should be understood as synonymous with the term "surfactant," which refers to any compound or mixture of compounds that have amphiphilic properties and reduce the surface tension of the liquid containing them. It is also believed that detergents may further enhance the removal of DNA from histones, thereby further aiding the efficiency of the process in inhibiting nuclease activity.

[0052] In certain embodiments, methods are provided in which the detergent is a quaternary ammonium compound detergent, for example, in situations where the sample is a plasma sample that may present problems due to the abundant presence of albumin. Quaternary ammonium compound detergents, such as cetyltrimethylammonium bromide (CTAB), have been observed to strongly promote albumin solubility and prevent albumin from saturating silica membranes. The quaternary nature of such detergents may also prevent them from altering the helical conformation of DNA, which may be advantageous, assuming they do not affect the binding efficiency of dsDNA to silica. During our experiments with challenging plasma samples, we appreciated the effectiveness achieved with CTAB, particularly because its effectiveness was observed even at very low silica binding concentrations between 0.25% and 1%. Such low concentrations are of interest for applications in hermetically integrated devices, where minimizing buffer volume is traded for maximizing sample input volume. Thus, in another embodiment, a method is provided wherein the quaternary ammonium compound detergent is cetyltrimethylammonium bromide (CTAB).

[0053] In another embodiment, the method is carried out within a fluidic cartridge, which may be a sealed fluidic cartridge that may form part of an automated system. In one embodiment of the above embodiment, the cartridge may be of a type suitable and adapted to directly receive a biological sample and obtain PCR-grade nucleic acids therefrom using the novel nucleic acid extraction chemistries presented herein, and to house at least one PCR reaction.

[0054] As used herein, the term "cartridge" should be understood as a self-contained assembly of chambers and / or channels formed as a single object that can be transported or moved as an attachment inside or outside a larger device suitable for receiving or connecting such cartridge. The cartridge and its device can be considered to form an automated system or platform. Some parts contained in the cartridge may be rigidly connected, while other parts may be flexibly connected and movable relative to other components of the cartridge. Similarly, as used herein, the term "fluidic cartridge" should be understood as a cartridge containing at least one chamber or channel suitable for treating, processing, discharging, or analyzing a fluid, primarily a liquid. An example of such a cartridge is shown in WO 2007 / 004103. Advantageously, the fluidic cartridge may be a microfluidic cartridge. In the context of a fluidic cartridge, the terms "downstream" and "upstream" may be defined relative to the direction of fluid flow in such a cartridge. That is, a section of a fluid path in a cartridge in which fluid flows toward a second section in the same cartridge should be interpreted as being located upstream of the second section. Similarly, a section where a fluid arrives later is located downstream relative to a section where said fluid passed earlier. Generally, as used herein, the terms "fluidic" or sometimes "microfluidic" refer to systems and devices that deal with the behavior, control, and manipulation of fluids that are geometrically constrained in at least one or two dimensions (e.g., width and height or channel) to a small, typically submillimeter, scale. Because such small volumes of fluid are moved, mixed, separated, or otherwise processed on a microscale, small size and low energy consumption are required.Microfluidic systems include structures such as micropneumatic systems (pressure sources, liquid pumps, microvalves, etc.) and microfluidic structures (microfluidic channels, etc.) that handle microliter, nanoliter, and picoliter volumes. Exemplary fluidic systems are described in EP 1896180, EP 1904234, and EP 2419705 and may therefore be applied in certain embodiments disclosed herein. In light of the above, the term "chamber" should be understood as any functionally defined compartment of any geometric shape within a fluidic or microfluidic assembly, defined by at least one wall and including the means necessary to perform the function attributed to said compartment. Along these lines, an "amplification chamber" should be understood as a compartment within a (micro)fluidic assembly that is suitable for and provided within said assembly for the purpose of performing nucleic acid amplification. Examples of amplification chambers include PCR chambers and qPCR chambers.

[0055] As used herein, the term "automated system" refers to an integrated platform containing equipment and disposable materials, such as plastics and solutions, that the system uses automatically to complete a specific process. While such a process may be initiated by a user, no user intervention is required throughout its automated processing within the system until the process is completed. As used herein, the term "instrument" refers to a machine equipped with at least a user interface (e.g., including at least a start button or electrical plug), an onboard computer with software, and programmed to perform certain functions, such as running an assay, which may include mixing, sonication, heating, data detection and collection, and, optionally, analysis. In possible embodiments, the interface may be in the form of a console with a computer system running user interface software that can initiate tests, display test results, and communicate with external information systems. A prominent automated system that can easily accommodate this method is the Idylla™ diagnostic platform manufactured by Biocartis NV, which uses disposable reagent-filled cartridges that can interface with cartridge processing equipment and provide sample-to-result analytical capabilities.

[0056] In alternative embodiments, there are further provided articles of manufacture directly related to and / or enabling the implementation of the above-described methods. In the simplest embodiment of such articles, a binding buffer solution is provided that contains both a buffering agent (e.g., acetate) adapted to maintain a pH between 3 and 6, and TMAC, both in concentrations directly adapted to achieve the desired concentration under the binding conditions of the silica solid support after mixing with the selected sample. Advantageous examples of such binding buffer solutions include, for example, 2.33 M TMAC and 0.47 M acetate, 3.6 M TMAC and 0.6 M acetate, 4.8 M TMAC and 0.8 M acetate, 6 M TMAC and 1 M acetate, 7.2 M TMAC and 1.2 M acetate, and 8.4 M TMAC and 1.4 M acetate.

[0057] In alternative embodiments of the above embodiments, a binding buffer solution is provided that further comprises a quaternary ammonium compound detergent and / or proteinase K at the appropriate concentrations described above. Using the example solutions listed above, the concentrations of CTAB can be 2.33 M TMAC, 0.47 M acetate, and 1.17% (wt / vol) CTAB, 3.6 M TMAC, 0.6 M acetate, and 1.5% (wt / vol) CTAB.

[0058] In alternative embodiments of the products disclosed herein, kits and / or fluidic cartridges containing any of the binding buffer solutions described above may be provided. As used herein, the term "kit" should be interpreted as a set of objects containing at least one item or assembly or item or equipment required for a specific purpose, such as performing a molecular biology process or assay. The kit may be provided in the form of a standard benchtop nucleic acid purification kit, including containers containing reagents such as binding buffers, wash buffers, etc., and one or more silica solid support spin columns, membranes, beads, etc. Alternatively, the kit may include a cartridge or simply be provided in cartridge form. Along these lines, in further embodiments, a cartridge is provided that contains a binding buffer solution, the buffer adapted to maintain a pH value between 3 and 6, and further contains TMAC. In further embodiments, the binding buffer solution in such a cartridge may further contain CTAB. In another embodiment, such a cartridge may advantageously further contain or include a silica solid support for nucleic acid purification. In further possible embodiments, such a cartridge may be a fluidic cartridge and / or may be adapted for processing liquid biopsy samples, such as plasma or urine.

[0059] Finally, also provided herein is the use of the methods and articles of manufacture (e.g., kits, cartridges, automated systems, etc.) described herein for the extraction of nucleic acids from liquid biopsy samples. In further embodiments, there is provided the use of the disclosed methods and articles of manufacture for the extraction of DNA, which is primarily double-stranded DNA (dsDNA), optionally cell-free DNA (cfDNA), or even circulating tumor DNA (ctDNA).

[0060] The novel nucleic acid purification chemistry and related products described herein may be applied to a wide variety of commercially available kits, lab-on-a-chip (LOC) devices, or disposable cartridges that use solid-phase extraction techniques to isolate nucleic acids from biological samples. More specifically, their use in fully integrated molecular diagnostic devices may be of great value, as the relatively small volume of binding buffer required allows for increased sample input relative to the required buffer volume. Examples of the concepts presented herein are provided below. [Example]

[0061] General experimental setup: A silica spin column (Machery-Nagel, blood column nucleospin) was attached to a QIAvac24 plus system, a vacuum manifold connected to a vacuum pump via a QIAvac connection system. The complete setup can be used as a flow-through system. The plasma sample and binding buffer solution were mixed in a 4 / 3 ratio (e.g., 1 mL of plasma and 0.75 mL of binding buffer) and applied to the silica spin column. Thus, the binding buffer was generally diluted 2.33-fold when mixed with the plasma. While the 4 / 3 ratio is not mandatory and is merely an arbitrary choice partially related to the design of the Idylla cartridge (the lysis chamber allows a maximum load of 7 mL), this is a silica spin column experiment. It is certainly possible to further increase the binding buffer concentration, thus reducing the required buffer volume relative to the sample volume. However, for this particular spin column setup, a 1.75-fold dilution of the plasma sample appeared sufficient in terms of clogging and flow rate. The silica membrane was subsequently washed with wash buffer, after which the spin column was removed from the vacuum manifold. The spin column was then placed in a 1 mL Lo-Bind Eppendorf tube and centrifuged at 10,000 revolutions per minute (rpm) for 1 minute. The spin column was then transferred to a new 1 mL Lo-Bind Eppendorf tube, after which elution buffer was added. After a 2-minute incubation at room temperature, the spin column was then subjected to an additional 1-minute centrifugation step at 10,000 rpm. The eluted product was then analyzed by qPCR to perform relative quantification of the purified DNA.

[0062] Sample type and binding buffer chemistry: Plasma (Innovative Research) was spiked with nucleosomal DNA (nDNA) isolated from whole blood. Spiking is useful for obtaining robust downstream qPCR-based target detection when processing smaller plasma volumes. Furthermore, nucleosomal DNA is characterized by a fragmentation pattern very similar to that of cell-free DNA (cfDNA). The presence of short fragments allows for evaluation of their extraction efficiency. 100 μL of plasma was spiked with 20,000 copies of nDNA. The spiked plasma was then mixed with 500 μL of binding buffer. The binding buffer consisted of 1.2 M tetramethylammonium chloride (TMAC) dissolved in 0.24 M sodium acetate (pH 5) buffer. This resulted in a final concentration of 1 M TMAC and 0.2 M sodium acetate when mixed with the plasma sample. This acidic mixture, with a total volume of 600 μL, was then loaded onto a silica spin column as described above.

[0063] Washing buffer chemistry: The silica membrane was washed by passing 1000 μL of 90% ethanol through the spin column, followed by a centrifugation step at 10,000 rpm for 1 minute to remove any residual ethanol traces.

[0064] Elution buffer chemistry: DNA elution was performed by rehydrating the silica membrane with water or Tris-HCl (pH 8.6) buffer. It is important that the elution buffer is at room temperature and in contact with the silica membrane for a minimum of 2 minutes. The spin column is then subjected to a final centrifugation step (1 minute, 10,000 rpm). The eluted product is then collected in a 1 mL Lo-Bind Eppendorf tube.

[0065] qPCR Design and Conditions: To evaluate the extraction efficiency of both short and long DNA fragments, a triplex design consisting of three amplicons of different sizes was used. The target amplicons were 62 bp, 98 bp, and 136 bp in length. The difference in Ct values between the shortest and longest amplicons indicated the presence of short target fragments. Primer and probe sequences can be prepared upon request. 20 μL of eluted product was mixed with 5 μL of PCR buffer. The final PCR reaction components were 50 mM KCl, 20 mM Tris-HCl (pH 8.6), 2 mM MgCl, 0.2 mM dNTP mix, 300 nM of each primer and probe, and 5 units of Gotaq DNA polymerase. qPCR reactions were performed in a Biorad CFX96 Touch™ Real-Time PCR Detection System. The total reaction volume was 25 μL. The cycling protocol included a hot start (95°C for 5 min) followed by 50 cycles of denaturation (95°C for 3 s) and annealing (64°C for 30 s). Fluorescence signals were measured after each cycle.

[0066] Results: First, we investigated the extraction efficiency of dsDNA in binding buffers with various chaotropic and kosmotropic or mild chaotropic salts at neutral pH. The PCR Ct values for 62 bp and 136 bp amplicons extracted in various binding buffers are shown on the Y-axis in Figure 1. The X-axis represents the composition of various binding buffers at neutral pH. "Input" is the reference point, representing the Ct value obtained when targeting the total amount of spiked nDNA. Thus, the delta Ct from the reference point indicates the extraction efficiency (i.e., a delta of 1 Ct = 50% extraction efficiency). If the delta Ct between the small and large amplicons remains the same as the reference point, this indicates no loss of small (62 bp to 136 bp) fragments. These results indicate that increasing the amount of kosmotropic salt (NaCl or (NH4)2SO4) in the binding buffer at neutral pH decreases the binding efficiency of dsDNA to silica. As explained previously, this is due to the presence of small kosmotropic cations (Na + and NH4 + We hypothesized that this may be due to the stabilizing effect of chaotropic salts on the DNA double helix. Apparently, under neutral conditions, high concentrations of chaotropic salts are preferred to mediate DNA binding to silica membranes.

[0067] The experiment was then repeated using acidic conditions (pH 5). The results are shown in Figure 2. As before, the Y-axis represents the Ct values for all three different amplicon sizes, while the X-axis represents the various binding buffer compositions. This data indicates that using acidic conditions (pH 5) and kosmotropic salts, as described in the prior art, does not efficiently mediate nDNA binding to silica membranes. 0.1 M NaCl at pH 5 appeared to perform best, with an extraction efficiency of approximately 6.25% (delta Ct = 4).

[0068] Next, the performance of the chaotropic Boom binding buffer (3.68 M GuSCN and butanol) was compared to a buffer containing chloride-based salts with and without quaternary ammonium compounds. The results are shown in Figure 3. The Y-axis represents the Ct values for all three different amplicon sizes. The X-axis represents the composition of the various binding buffers. The data are based on the quaternary ammonium cations (TMA). + ) and kosmotropic anions (Cl - ) efficiently mediates DNA binding to silica membranes. These results support the hypothesis that the inertness of quaternary cations allows them to destabilize the helical structure of DNA, thereby increasing the amount of available silica binding sites.

[0069] Next, we investigated the optimal pH range for plasma samples. The results are shown in Figure 4. The Y-axis represents the Ct values for all three different amplicon sizes. The X-axis represents the composition of the various binding buffers. This experiment demonstrates that lowering the pH of the binding buffer to 4 is incompatible with native plasma samples. As soon as 0.3 M TMAC is added, protein aggregation becomes so severe that successful sample processing becomes nearly impossible. This is most likely related to the isoelectric point (pI) of albumin (4.7), which is abundant in plasma. As soon as the solution pH approaches the protein's pI, charge repulsion between individual protein molecules decreases, potentially leading to precipitation. At this point, it appears that even a slight dehydrating effect by anions is sufficient to promote protein aggregation.

[0070] As a next step, the performance of TMA sulfate (TMAS) was compared with that of TMAC, and the results are shown in Figure 5. The Y-axis represents the Ct values for all three different amplicon sizes. The X-axis represents the composition of the various binding buffers. This experiment demonstrates that TMAS does not efficiently mediate DNA binding to silica. These results suggest that the primary relevant binding mechanism is (i) Dehydration of the silica membrane and nucleic acid, achieved by providing sufficient kosmotropic anions, thus reducing the amount of free water; (ii) shielding of intermolecular electrostatic forces, achieved by using acidic conditions to protonate negatively charged silanol groups; This was initially surprising based on our initial hypothesis that

[0071] Based on the above, it was expected that sulfate ions, due to their double charge, would be more kosmotropic than chloride ions and thus exhibit a stronger dehydrating effect, leading to more efficient binding of DNA to silica. As confirmed in this and many other experiments, it was concluded that mechanisms other than those described above must be applicable. Non-Patent Document 1 described (iii) the formation of intermolecular hydrogen bonds at the nucleic acid-silica contact layer as a third effect that could affect DNA-silica interactions. This data appears to suggest that these hydrogen bonds may be more important than initially thought, and that the use of strongly kosmotropic sulfate anions strongly disrupts or even prevents the formation of these bonds. Therefore, the performance of chloride anions, specifically described as weakly kosmotropic or bordering on kosmotropic / chaotropic, may be inherent in their specific nature, in that their interactions with water are not stronger than their interactions with water itself.

[0072] To further investigate this hypothesis, we compared the performance of various TMA-containing salts in mediating the binding of spiked nDNA to silica in both PBS and plasma samples. The results are shown in Figure 6. The Y-axis represents the Ct value of the 62-bp amplicon. The X-axis represents the composition of various binding buffers. "Input" indicates the reference point, representing the Ct value obtained when targeting the total amount of spiked nDNA. Thus, the delta Ct from the reference point indicates the extraction efficiency (i.e., a delta of 1 Ct = 50% extraction efficiency). This data demonstrates the importance of anion selection. The charge density of chloride ions allows for the highest extraction efficiency at the lowest concentration. More kosmotropic anions, such as sulfate ions, which have a higher charge density, rapidly decrease extraction efficiency, possibly by mediating a conformational change from B-DNA to A-DNA or by other unknown mechanisms. On the other hand, more chaotropic anions such as bromide, which have a lower charge density, are likely to prevent such conformational changes, but are significantly less efficient at dehydrating silica membranes, requiring higher molar concentrations to achieve performance comparable to that provided by the selection of chloride. It should also be noted that the reduced charge density of more chaotropic anions also adversely affects the solubility of quaternary ammonium salts. From this perspective, TMAC is superior in both performance and solubility.

[0073] Having identified TMAC as the most promising quaternary ammonium salt, we next investigated its optimal concentration in our specific experimental setup. The results are shown in Figure 7. The Y-axis represents the Ct values for all three different amplicon sizes, while the X-axis represents the composition of the various binding buffers. These results indicate that increasing the concentration of TMAC in the binding buffer above 1 M does not have a beneficial effect on the efficiency of DNA binding to silica. In fact, increasing the TMAC concentration slightly decreases the binding efficiency.

[0074] Next, we investigated the dsDNA extraction performance using TMAC at various pH values. The results are shown in Figure 8. The Y-axis represents the Ct values for all three different amplicon sizes. The X-axis represents the composition of various binding buffers. These results demonstrate the importance of the pH of the binding buffer. As previously explained, when processing undigested plasma, lowering the pH of the binding buffer to near the isoelectric point of albumin (4.7) results in severe protein aggregation, making it impossible to process the sample in a spin column or microfluidic channel. Furthermore, increasing the pH of the binding buffer above 5 slightly increases the negative charge repulsion between the silanol groups and the phosphate backbone of DNA, resulting in a decrease in the efficiency of DNA binding to silica. Surface silanol groups on silica membranes are known to have pKa values ranging from 4 to 8. Increasing the pH likely leads to deprotonation of the silanol groups with the lowest pKa values, which make them negatively charged.

[0075] Next, we investigated the TMAC molarity and acetate buffer pH more extensively for multiple different plasma batches. Furthermore, we also added the quaternary ammonium detergent cetyltrimethylammonium bromide (CTAB) to the binding buffer. The results are shown in Figure 9. The Y-axis represents the Cq value of the 62-bp amplicon. The X-axis represents the binding conditions, including various amounts of added CTAB. Unlike previous experiments, in which a limited amount of plasma was processed after spiking with nDNA, this experiment focuses on extracting cfDNA from 1 mL of unspiked plasma. For further improvement, plasma samples were also subjected to a protein digestion step using 1 mg / mL proteinase K at 37°C for 10 min. Each 1 mL plasma sample was treated by adding 0.75 mL of binding buffer (2.33 M TMAC, 0.47 M acetate, 1.17% CTAB, pH 5). Subsequently, the membrane was washed with 1 mL of the first wash buffer (1 M TMAC, 0.2 M acetate, pH 5), and finally, the membrane was washed with an additional 1 mL of 90% EtOH. Ten different samples were processed for each binding condition. The boxplots in Figure 9 show the mean and median Cq values and the variability. It is clear that the addition of CTAB can have a beneficial effect, as it can reduce sample-to-sample variability and increase DNA yield. This effect is likely caused by CTAB promoting albumin solubility and thus preventing albumin deposition on the silica membrane. It also likely inhibits nuclease activity, possibly aiding in the removal of cfDNA from histones as well.

[0076] Next, we looked at different TMAC molarities at different pH values. By evaluating different binding conditions on multiple plasma batches, we found that different samples responded very differently to different conditions, as shown in Figure 10. The Y-axis shows the Ct value of the 62 bp amplicon. The X-axis represents the different pH values of the binding conditions. The shape of the plot represents the different TMAC molarities (as explained in the legend).

[0077] Next, the efficiency of the elution conditions was investigated. The results are shown in Figure 11. The Y-axis shows the Ct values for all three different amplicon sizes. The X-axis shows the various elution buffer compositions and incubation times at room temperature. The data show that the elution step is most efficient when a slightly basic buffer is used. Again, this is likely related to the pKa value of the surface silanol groups. Negative charge repulsion between the silanol groups and the phosphate backbone is thought to be the main driving force during elution with rehydration. Therefore, providing an elution buffer with a pH above the highest pKa value of silanols (8) ensures that all silanol groups are negatively charged. Figure 11 shows that a K of 50 mM + and 2 mM Mg ++ It can be seen that the presence of cations negatively impacts the elution efficiency. It is hypothesized that this effect is likely caused by these highly kosmotropic cations shielding the negative charge repulsion between the silanol groups and the phosphate backbone. This observation should be kept in mind when adapting this protocol to fully integrated molecular diagnostic devices where silica-bound nucleic acids are eluted directly with amplification buffer.

[0078] Next, we investigated the potential beneficial effect of including a proteinase K predigestion on the volume of starting material, particularly for challenging plasma samples. The results are shown in Figure 12. The Y-axis represents the Ct value of the 62-bp amplicon. The X-axis indicates the various plasma volumes processed. These results demonstrate that including a proteinase K digestion step increases DNA yield when processing plasma samples greater than 400 μL. Plasma was digested at 56°C for 10 minutes. As investigated herein, protein binding to silica membranes may also occur, in part, under the chemistry of the binding buffer. Acidic conditions reduce the negative charge of albumin, which has a pI of 4.7, thereby significantly reducing charge repulsion between this abundant plasma protein and the silica membrane. Thus, proteins and nucleic acids are expected to compete for binding sites, although the surface of the silica membrane remains limited. Furthermore, reduced charge repulsion between individual protein molecules allows these protein molecules to overlap much more closely when bound to the membrane. Therefore, digestion or removal of albumin from plasma is likely to be beneficial to improve binding of nucleic acids to silica membranes when processing larger samples.

[0079] Next, we evaluated the scale-up of plasma sample volumes in Biocartis NV's proprietary disposable cartridges that accompany the Idylla™ integrated system. The results are shown in Figure 13. The Y-axis represents the Ct values of the 62 bp amplicon. The X-axis represents the various plasma volumes processed using various binding buffer chemistries. The samples used were unspiked plasma samples. These results are encouraging, demonstrating that the novel binding buffer chemistry presented herein allows for significant sample scale-up in the Idylla cartridge, resulting in a linear increase in yield. The binding conditions applied were: 1.2 M TMAC, 0.2 M acetate, 0.5% CTAB, pH 5. The final buffer composition used on the cartridge was as follows: Binding buffer (3 mL): 2.8 M TMAC, 0.47 M acetate, 1.17% CTAB, pH 5 (diluted 2.33-fold with sample), Plasma samples (4 mL) (treated with 1 mg / mL proteinase K for 10 minutes at room temperature), First wash buffer (1.25 mL): 1.2 M TMAC, 0.2 M acetate, pH 5, Second wash buffer (2.4 mL): 90% ethanol, Elution buffer: HO (volumes to suit any requirement, for Idylla the minimum elution volume is 160 μL and the maximum elution volume is 250 μL).

[0080] Finally, we compared the performance of cartridges using the disclosed novel extraction chemistry with a chaotropic reference chemistry. Seven different plasma batches were used, with five cartridge replicates per batch for each extraction chemistry. All 70 cartridge runs were completed successfully without any clogging errors. The results are shown in Figures 14 and 15. In Figure 14, the Y-axis shows the Ct value of the 62-bp target amplicon. The X-axis specifies the extraction chemistry (chaotropic reference on the left and novel TMAC+CTAB chemistry on the right for each panel) and sample volume (1 ml and 4 ml, respectively). Each panel represents a different plasma batch. In Figure 15, the Y-axis shows the Ct value for the same amplicon, while the X-axis lists the different plasma batches. Each panel represents a different extraction chemistry (chaotropic reference on the left and novel TMAC+CTAB chemistry on the right). The open dots are a 10-fold dilution of the solid dots and are therefore indicative of PCR inhibition. The data in Figures 14 and 15 demonstrate that the increased sample input (4 ml) made possible by the use of the new extraction chemistry results in a substantial increase in yield. The actual measured yield increases are listed in Table 1 below. Figure 15 further demonstrates that none of the sample extracts contained PCR-inhibiting components. These results are robust, demonstrating that the new binding chemistry has a standard deviation comparable to that of the chaotropic reference chemistry. Furthermore, as shown in Table 1, for all plasma batches studied, at least a two-fold increase in yield was obtained thanks to the ability to increase the sample input volume per cartridge. In addition to this increased assay sensitivity, it should be noted that the new buffer chemistry is completely chaotrope-free and low-cost.

[0081] Table 1: Average Ct values obtained for each batch of plasma in Figure 14. The delta Ct between both extraction chemistries reflects the increase in cfDNA yield. [Table 1]

[0082] Drawing translation Figure 1 100 μL sample + 500 μL buffer 100 μL sample + 500 μL buffer 200 cps / μL nDNA 200 cps / μL nDNA Plasma input input water Figure 2 100 μL sample + 500 μL buffer 100 μL sample + 500 μL buffer 200 cps / μL nDNA 200 cps / μL nDNA Plasma input input Figure 3 200 cps / μL nDNA 200 cps / μL nDNA 100 μL sample + 500 μL buffer 100 μL sample + 500 μL buffer Plasma input input Figure 4 Plasma 100 μL sample + 500 μL buffer 100 μL sample + 500 μL buffer 200 cps / μL nDNA 200 cps / μL nDNA input input Figure 5 undiluted 100 μL sample + 500 μL buffer 100 μL sample + 500 μL buffer 200 cps / μL nDNA 200 cps / μL nDNA Plasma input input Figure 6 100 μL sample + 600 μL buffer 100 μL sample + 600 μL buffer plasma input input water nDNA input Figure 7 100 μL sample + 500 μL buffer 100 μL sample + 500 μL buffer 200 cps / μL nDNA 200 cps / μL nDNA Plasma input input Figure 8 200 cps / μL nDNA 200 cps / μL nDNA 100 μL sample + 500 μL buffer 100 μL sample + 500 μL buffer Plasma input input Figure 9 1mL plasma 1 mg / mL Qiagen protK 10' 37°C 1 mg / mL Qiagen proteinase K 10 min at 37°C undiluted 10-fold dilution Average Average Median Figure 10 1.0 mg protK / Qiagen 10' 37℃ 1.0 mg proteinase K (Qiagen) 10 min at 37℃ 1mL plasma 3.7M GuSCN & 43% ButOH 3.7 M GuSCN and 43% ButOH Figure 11 100 μL sample + 500 μL buffer 100 μL sample + 500 μL buffer 200 cps / μL nDNA 200 cps / μL nDNA Plasma Water 0' 0 minutes 2' 2 minutes 5' 5 minutes Figure 12 200 cps / μL nDNA 200 cps / μL nDNA Plasma Native plasma Proteinase K digested plasma Figure 13 10-fold diluted undiluted 1mL plasma 2mL plasma 4mL plasma 3.7M GuSCN & 43% ButOH 3.7 M GuSCN and 43% ButOH 1.2M TMAC 0.2MA acetate 0.5% CTAB pH5 1.2 M TMAC, 0.2 M acetate, 0.5% CTAB, pH5 Figure 14 Chaotropic chemistry CTAB chemistry Count Avg Average StdDev standard deviation Figure 15 Chaotropic chemistry CTAB chemistry 10-fold diluted undiluted

Claims

1. The liquid biopsy sample and the silica solid support are mixed at a pH value between 3 and 6, and small quaternary organic compounds, defined as quaternary compounds consisting of a central positively charged atom bearing four organic substituents R1-R4, where the number of carbon atoms in each of the organic substituents R1-R4 does not exceed two; a bromide anion or a chloride anion; A method for extracting dsDNA, comprising contacting in the presence of a salt consisting of: The method wherein the positively charged atom of the small quaternary organic compound is nitrogen, and the concentration of the small quaternary organic compound is between 0.1 M and 2 M.

2. The method of claim 1 , wherein the anion is a chloride ion.

3. 3. The method of claim 2, wherein the small quaternary organic compound is tetramethylammonium chloride, further referred to as TMAC.

4. 4. The method of claim 1, wherein the concentration of the small quaternary organic compound is 1.2 M.

5. The method according to any one of claims 1 to 4, wherein the pH value is between 4 and 5.

8.

6. The method according to any one of claims 1 to 5, wherein the pH value is between 4.8 and 5.

6.

7. The method according to any one of claims 1 to 5, wherein the pH value is between 4.6 and 5.

4.

8. The method according to any one of claims 1 to 7, wherein the pH value is 5.

9. The method according to any one of claims 1 to 8, wherein the method is preceded by a protease treatment.

10. The method of any one of claims 1 to 9, wherein the liquid biopsy sample is selected from plasma, serum, whole blood, or urine.

11. The method of any one of claims 1 to 10, wherein the dsDNA is cell-free dsDNA.

12. 12. The method of claim 11, wherein the dsDNA is circulating tumor dsDNA.

13. The method of any one of claims 1 to 12, wherein the contacting is carried out in the presence of a detergent.

14. 14. The method of claim 13, wherein the detergent is a quaternary ammonium compound detergent.

15. 15. The method of claim 14, wherein the quaternary ammonium compound detergent is cetyltrimethylammonium bromide, further known as CTAB.

16. The method according to any one of claims 1 to 15, wherein the method is carried out inside a cartridge.

Images