CA2082186A1 - Production of high binding capacity magnetizable particles of controlled size used in immunoassay systems - Google Patents

Abstract

Abstract High binding capacity magnetized particles, which represent a significant improvement over the existing paramagnetic particles, were developed. It was found that the use of high pH silanization yields particles that maintain the original size throughout the manufacturing process. These exhibit good binding capacity and do not require the use of hazardous solvents during manufacture. Also, it was found to be a more efficient process, in that an 80 - 90%
reduction in the use of silane was realized.

Description

2~2~8~ ~

PRODUC'~ION OF HIGH BINDING CAPACITY MAGNETIZABLE PARTIC~ES
OF CONTROLLED SIZE USED IN IM~NOASSAY SYSTEMS

Backaround The use of magnetic particles for biochemical separations has been known for some time. (U.S. Patents ~,554,088;
4,628,037; ~,695,392; 4,695,393; Hersh, L.S., Yaverbaum, S., Magnetic Solid-Phase Radioimmunoassay, 63 Clin. Chim. Acta (1975) 69; Pourfarzaneh, M., Johnson, S.C., Landon, J., Production and Use of ~agnetizable Particles in Immunoassay, 5 The ~igand Quarterly (~82~ 41.) They are particularly use~ul in the separation of immwlochemical complexes, in which one of the members of the immunochemical pair (i.e., either the antigen or antibody) is coupled to the magnetic particle. When this parti¢le complexes to the other member of the i~munochemical pair, the complex is then easily separated from the solution by the use of a ma~net.

The Whitehead patents co~er the coatPd iron oxide particles :~
themselves (U.S~ Patent No. 4,695,392), the particles coupled to a ~ioaffinity adsorbent ~U.S. Patent No~
4,S54,088), the use of the particles for determining the concentration of a ligate (U.S. Patent No. 4,628,037~, and the process for preparing the magnetically-responsive particles ~V~SO Patent No. 4,69S,393).
: ' The existing (Whitehead) process for preparing these 2~82:~g~

particles involves the use of acidic pH conditions (pH 4.5), to which a 10~ silane amine solution is added, with the mixture heated at 90 - ~5 QC for about 2 hours followed by a glycerol dehydration step and a wash of the final product with copious amounts of methanol and water. This results in a decrease in particle size and a corresponding decrease in sedimentation rate for the particles. ln addition, the existing process involves the use of hazardous solvents (e.g., methanol) and, therefore, requires the use of special facilities and equipment (to reduce the risk from possible explosion) and the need to take precautions regarding the disposal or recycling of the spent solvent.

Summary High binding capacity magnetized particles, which represent a significant improvament over the existing paramagnetic particles, were developed. It was found that the use of high pH silanization yields particles that maintain the original size throughout the manufacturing process. These exhi~it good binding capacity and do not require the use of hazardous solvents during manufacture. Also, it was found to be a more efficient process, in that an 80 - 90%
reduction in the use of silane was realized.

Details of the Invention This invention rslates to an improvement in the process for making paramagnetic particles which are useful in ~;
biochemical separations. One major use is in immunochemical assays, wherein a biochemically active species (e.g., an antibody) is chemically bound to the paramagnetic particle.
After the paramagnetic particle-antibody complex has reacted with any antigen in solution, the presence of the paramagnetic material simplifies the separation of the - . ~

2~2~

antigell-antibody complex for further analysis. Other applications, for example, in affinity purification chromatography, are known for these paramagnetic particles by those experienced in the art.

Although paramagnetic particles made of iron salts are prefPrred, particles made from any material or combination of materials having magnetic properties can be used in the novel process for making coated magnetic particles. For example, chromium salts or mixtures of iron salts and glass can be used in the process.

It was found that, by conducting the silanization process in a high pH environment, an unexpected benefit was found, in that relatively high binding capacity was obtained and maintained throughout the manufacturing process. As will be further discussed hereafter, the high binding capacity particles provide improved performance in the immunochemical assay. In addition, since the particles are washed using water, instead of a flammable solvent, there is an improvement in the safety and wa~e disposal aspects of making the particles. Furthermore, the manufacturing efficiency and cost aspects are improved, since considerably ~-less silane (approximately one-fifth as much) is needed to manufacture the particles.

One discovery regarding the paramagnetic particles related to the fact that both the sodium hydroxide concentration and total concentration of ferrous chloride and ferric chloride used during the preparation of iron particles could affect particle size. Figure l, which shows particle siz as a function of total iron concentration at two levels of sodium hydroxide, shows that the use of higher NaOH concentration ~21~

would lead to larger iron particles. The particle size was determined by measuring the sedimentation rate for the particles. Throughout the experiment reported in Figure 1, the molar ratio of ferrous to ferric ions remained at 2:1, but the total iron concentration was varied. The results showed that as total iron concentration increased, particle size, as determined by sedimentation rate, increased. The variation of NaOH concentration was used to prepare iron core particles of desired siæe that would be used in the next step of the process, the silanization step.

It was also learned t~at the novel process was more efficient in silane utilization, in that a much higher concentration of silane was found to be deposited on the magnetic particles. Particles silanized at pH 11.0 were compared to those prepared using the former technique ~where the iron particles were prepared a~ pH 5.0), and the pH 11.0 - prepared particles were found to contain at :Least 5 times the silane content as those prepared at pH 5Ø (See Figure 2, which shows free amine content as a function of time, at two pH~s~ For the data in Figure 2, the SPDP method, to be discussed in more detail hereafter, was used to determine amine concentration on the particles. It should be noted ;~
that amine content was used as an indicator of silane content.) Thus to obtain equivalent silane content, much less silane is needed at pH 11Ø

The effect of p~ on silane amine deposit on the particles was determined by measuring the amount of stable primary amine on the particle. This was done by a new analytical technique using N-succinimidyl-3-(2-pyridyldithio3propionate (or SPDP), discussed in detail in Example 3. In the immunochemical analysis in which the magnetic particles are . . , 2~82~g~

used, it is advantageous for the particles to have more silane and, subsequently, more biologically active materials could be bound This higher capacity particle (shown vs.
reference material in Figure 2) was, there~ore, selected for use in the silanization, and, as can be seen in Figure 2, this resulted in more silane amines being immobilized on the particles.

In addition, the novel process represented an improvement in safety and environmental aspects because of the ability of using water as the solvent, as opposed to methanol in the previous manufacture technique.

The ratio of ferrous to ferric could vary, but preferably the ratio was equal to one or was greater than one.
Somewhat more preferred was the ratio of 1.0 to 3Ø Most preferably the ratio was equal to 2. At this ratio, the particle size was found to be dependent on the molarity of the NaOH used. It was also found that the strong magnetic property correlated to the color of the particle~. When the particles were ~lack, the strongest magnetic properties were 0 found, and the black color was found when the ratio of ~0--ferrous to ferrlc~exceeded one. (See Table 1) ~4~ ~
The silane that can be used to coat the particles can be any -' silane with one or more reactive groups per molecule. The general chemical structure of the preferred silane compound can be depicted as OR OR
Ro-si-~cH2)p~x or Ro-si- (CH2) q-Y- (CH2) p-X
OR OR
where R = methyl, ethyl, or propyl, and the R groups do not -~ 2~82~ 8~ ~

need to be the same in the same molecule. p can range ~rom 1 to 5 and q can rang~ from O to 5. X can be NH~, SH, Cl, ~o /01 CN, , -CH2-CH-CH2, or -O-CO-C~CH3)=CH2, and Y
can be NH, benzyl sr substituted benzyl, or oxygen.
Preerably the silane compound is one where p is between 1 and 3, where the R groups are the same within one molecule and where X is an amino group. The most preferred silanes are alkylamine silanes, such as N-2-aminoethyl-3-aminopropyltrimethoxysilane, 3-aminopropyltrimethoxysilane, n-hexyltrimethoxylsilane, and n-dodecyltriethoxysilane. -The following examples describe various aspects of the preparation, analysis and use of the magnetizable particles described herein but are not intended to limit the usefulness of the newly invented materials or processes.

PREPARATION OF MAGNETIZABLE PARTICLES

A laboratory batch size of approximately 4 - 5 grams of particles were prepared. 98 ml of 0.5 M ferrous chloride and 9-8 ml of 0.25 M ferric chloride were mixed with 49 ml of distilled water in a 500 ml erlenmeyer flask. 49 ml of 5 M

NaOH pr~-warmed to 60 C. was poured into the iron chloride :. , .
~: . , - , , , ~

~$~

mixture while swirling. Black iron oxide particles were ormed instantly. The particles were precipitated by placing a strong magnet under the erlenmeyer, and the supernatent was decanted. The particles were then washed with 200 ml of an aqueous solution containing 0O05 M NaCl, and the supernatent was decanted and its pH measured. The NaCl washings were continued until the pH of the supernatent reached neutrality.

EX~MPLE 2 SILANIZATION OF THE PARTICLES

In a 500 ml, 3-arm, round bottom flask, 4 - 5 grams of the magnetizable particles from Example 1 were suspended in 150 ml of aqueous solution containing 2% by weight of alkylamine trimethoxy silane and 0~05 M NaCl. The pH of the mixture should be llo0~ A cooling condenser was placed in one side arm, a nitrogen gas intake was placed in the other side arm, ~ -while a paddle-stirrer was placed in the center arm. The flask was immersed half-way in a silicone oil bath at 90 -95 C. The reaction is allowed to proceed for 2 hours, with nitrogen set at 0.1 pounds per square inch pressure and with the stirrer rotating slowly. After the reaction was stopped, the particles were washed 5 times with 100 - 150 ml of an aqueous solution containing 0.05 M NaCl. The 2~2~

particles were then suspended in loO ml of 0.05 M NaCl solution and were ready f~r amine determination by the SPDP
method.

~XAMPLE 3 N-succinimidyl-3-(2-pyridyldithio)propionate (or SPDP) is used in a novel technique to determine amin~ content of the silanized particles. Duplicate "unknown" samples were run vs. duplicate non-silanized samples (controls), which were used to determine background readings. 3 - 4 mg of parti~les were washed with distilled water for 6 times and were resuspended in 1 ml of distilled water .in a 12 x 75 mm glass test tube. Duplicate samples were heated in a 100 C.
water bath ~or 0, 15, 30 and 60 minutes. All samples were washed again twie with water~ then s~spended in 1 ml of 0~1 M sodium phosphate buffer/0.15 M NaCl, at pH 8Ø Each sample was then transferred quantitatively into a 18x15Q mm glass test tube, to which is added 100 ~l of a solution of SPDP (5 mg/ml in ethanol). The samples are mixed well and -shaken at room temperature for 1 hour. Each sample was then washed 3 times with 1 ml of 0.01 M phosphate buffered saline (PBS) at pH 7.4 and then resuspended in 2 ml of the same PBS
buffer. To each sample was added 50 ~l Gf dithiothriotol '' ' ' -, ' .

' 2~21~

_9 _ solutio~ (10 mg/ml in the PBS buffer). 2-thiopyridine is then released quantitatively. tSee Figure 3.~ Nano-equivalent concentration of amine is calculated bas~d on the standard curve of 2-thiopyridine absorbance at 343 nm.

PARTICLE SIZE DETERMINATION

Particle size was determined by sedimentation kinetics in water, with the faster sedimentation rates indicating larger particle size. The sedimentation was determined by observing the change in optical density at 600 nm as a function of time. Optical density chanye over a period of about 4 ~inutes was measured, and the rate was calculated to be the change in O.D./time. Control lots made by the former method were used as controls. Typical data are sho~n in Figure 1.

PREPARATION OF MAGNETIC PARTICLES COUPLED TO ANTIBODY

To illustrate the preparation of magnetic particles coupled to antibody, the immobilization of antibody against thyroid stimulating hormone (TSH) is described.

~2~

110 mg of silanized particles were first heat treated at 100 C. for 30 minutes to remove unstable amine. They were then activated with 20 ml of 10~ glutaraldehyde in 0.1 M
sodium phosphate buffer, at p~ 7.4~ for 3 hours at room temperature. The particles were then washed three times with the same buffer and then resuspended in 10 ml of the same buffer containing 2~ mg of anti-rrSH immunoglo~ulin~
After mixing on a rocker overnight ~t room temperature, the supernatant was decanted, and the particles were suspended 10 in 10 ml of a solution containing 10 mg/ml of bovine serum albumin tBSA) in 0.1 M phosphate buffer, pH 7.4. ~he suspension was placed in a 50 C. water bath and shaken for 4 hours. Afte.r 3 washes with 0.1 M ~odium phosphate buf~er, pH 7.4, the particles were resuspended in 50 ml of the PBS/BSA buffer (0.01 M sodium phosphate, 0.15 M NaCl, 1 mg/ml bovine serum albumin, pH 7.4) and evaluated in the TSH
assay system.

USE OF COUPLED MAGNETIC PARTICLES IN IMMUNOCHEMICAL ASSAY

The coupled anti-TSH particles prepared in Example 5 were analyzed by their use in an immunochemical assay. ~uplicate samples were run and averaged to obtain each value.

.

':' : . , : ~ ' .
., ' - : . .

2~18~

loo ml of a sample containing a known amount of TSH was mixed in a flask with 100 ~1 of the coupled antibody prepared in Example 5 which was labeled with I~I. After 2 hours of incubation at room temperature, 500~1 of the solid phase antibody was added. The solution was stirred and then incubated further for 30 minutes at room tamperature. Then the solid phase was separated ~y placing a strong magnet under the flask. After the supernatent was decanted, the remaining particles were washed with 1 ml of water, and, after exposing to a magnet for 3 minutes, the supernatant was again decanted. The residue was then counted using radioisotope techniques. The same assay was run using the particles made by the foxmer manufacturing technique (PMP) as a control. ~he results are shown in Table 2. It was found that the particles made by the novel technigue ~HMP) gave lower non-specific binding. Bo~h sensitivity and slope of the standard curve using the novel technique were approximately doubled, the improvement being attributed to the higher amine content which led to improved coupling efficiency.

~4382.~g EFFECT OF FE~/FE~+
ON YIELD, PARTICLE SIZE, PARAMAG~ETISM, COLOR

RATIO YIELD SEDIMENTATION RATE PACKED COLOR
MIM ~lm~ In H~ In NaCl Solu OD~MIN VOL. RBL. DRY

1:1 80 4 5 1303E-3 2 BLK
1:2 35.6 5 4 1.25E-3 1 BRN
1:3 10.5 3 3 6.7E-3 3 BRN
2:1 68.3 1 1 12.5E-3 4 BLK
2:3 80.4 3 4 3.75E-3 3 BRN

3:1 57.5 2 3 6.2.5E-3 3 BLK
3:2 68.0 2 2 6.25E-3 3 BLK

Relative values, with 2-1 set at l All wet cakes appear to be black. When dried, they appear either black or brown in color~
Significant particle loss during washing steps 2~2.1l~ ~

COMPARISON OF RIA TSH SENSITIVITY BETWEEN PMP AND HMP
AS SOLID SUPPORT FOR TSH ANTIBODY

STANDARD
_8~ L_ ~ 2~1Ekl HMPL2MG/ML) HMP(lMG/ML~. HMPt.5MGLML) 0.751,01~ 1,031 803 611 2 1,952 1,96~ 1,535 1,138 6 4,264 4,901 3,667 2,610 ~,443 9,245 7,248 5,410 3~ 13,668 15,12~ 11,865 8,~89 60 22,023 23,619 19,132 13,738 SENSITIVITY
2.97 4.10 6.64 6.7 MAX. SIGNAL/
NOISE R~TIO
64.21 94.10 15~.12 152.64 (All values are means of duplicate samples. No overlappingg readings between two adjacent standards.) PMP are particles made at pH5. HNP are particles made at pHll.
As measured by reading of 0.75 ~IU standard vs. that of 0.
As measured by reading of 60 ~IU standard vs. that of 0.

Claims (16)

1. A process for making a coated magnetic particle comprising reacting a suspension of the particles with silane in alkaline pH conditions.

2. The process of claim 1 wherein the particles are iron oxide particles.

3. The process of claim 1 wherein the alkaline pH is approximately pH 11.

4. The process of claim 2 in which the ratio of ferrous to ferric particles is equal to or greater than l.

5. The process of claim 4 in which the ratio is equal to or greater than 1.0 and less than 3Ø

6. The process of claim 5 in which the ratio is equal to 2.

7. The process of claim 1 in which the silane has the formula or where R = methyl, ethyl, or propyl, and the R groups do not need to be the same in the same molecule, p can range from 1 to 5, q can range from 0 to 5, and X can be NH2, SH, Cl, CN, , , - O-CO -C(CH3)=CH2.

8. The process of claim 7 where p is between 1 and 3, the R groups are the same within 1 molecule, and X is NH2.

9. The process of claim 8 in which the silane is an alkylamine silane.

10. The process of claim 9 in which the silane is N-2-aminoethyl-3-aminopropyltrimethoxysilane.

11. The particles produced by the process of claim 1.

12. The process of using the particles of claim 11 in an immunochemical assay.

13. The process of claim 12 which comprises the coupling of the particles with a bioaffinity adsorbent prior to its use in the immunochemical assay.

14. The process of claim 13 in which the bioaffinity adsorbent is either an antigen or antibody.

15. A method for analysis of stable amines comprising the use of N-succinimidyl-3-(2-pyridyldithio)propionate.

16. A method of claim 15 comprising (a) reacting the amine with N-succinimidyl-3-(2-pyridyldithio)propionate, (b) reacting the isolated product from step (a) with dithiothreitol, thus releasing 2-thiopyridine, and (c) comparing the absorbance of the solution resulting from step (b) with a standardized curve to determine the concentration of amine.