I am writing to reflect a laboratory experiment I performed based on the publication A Simplified Method for the Analysis of Hydroxyproline in Biological Tissues written by G. Kesava Reddy and Chukuka S. Enwemeka.

Abstract:

"A critical study of the different steps involved in previous procedure for hydroxyproline assay allows the direct measurement of collagen content in tissue homogenates without losing the advantages of the method.  The procedure is based on alkaline hydrolysis of the tissue homogenate and subsequent determination of the free hydroxyproline in hydrolyzates.  Chloramine-T was used to oxidize the free hydroxyproline for the production of a pyrrole.  The addition of Ehrlich's reagent resulted in the formation of a chromophore that can be measured at 550 nm.  Optimal assay conditions were determined using tissue homogenate and purified acid soluble collagen along with standard hydroxyproline.  Critical parameters such as the amount of Chloramine-T, sodium hydroxide, p-dimethylaminobenzaldehyde, pH of the reaction buffer, and length of oxidation time were examined to obtain satisfactory results.  The method has been applied to samples of tissue homogenate and purified acid soluble collagen, with recovery of added hydroxyproline of 101 +/- 6.5 and 104 +/- 6.0 (SD) percent, respectively.  The method is highly sensitive and reproducible when used to measure the imino acid in tissue homogenates.  The modified hydroxyproline assay presented in this communication will be useful for routine measurement of collagen content in extracts of various tissue specimens.  In addition, the modified method can be used for batch processing of column fractions to monitor the collagen concentrations during purification."

This experiment is restricted to observe absorbance values for different amounts of hydroxyprolene.

Experiment:

Preparation of Reagents:

Hydroxyprolene stock:  0.500 g of hydroxyprolene was measured and prepared in 1 L solution of de-ionized water.

Acetate-citrate buffer pH 6.5: 120 g of sodium acetate trihydrate, 46 g of citric acid, 12 mL acetic acid, and 34 g sodium hydroxide pellets were dissolved in distilled water.  The pH measured out to be around 11, but it was brought down to 6.5 with further addition of glacial acetic acid.  The solution was then brought to one liter by adding water.

Chloromine-T reagent (0.056M): An old stock of Chloromine-T failed the flame test for the existence of chlorine, so a fresh stock was purchased.  1.27 g of fresh Chloramine-T was dissolved in 20 mL 50% n-propanol and brought to 100 mL with acetate-citrate buffer.

Ehrlich's Reagent (1M): 1.5 g of p-dimethylaminobenzaldehyde was disolved in 6.7 mL n-propanol/3.3 mL perchloric acid and brought to 10 mL.  This reagent is not stable, and must be prepared for each assay series.

Assay Procedure I:

1.  One blank (50 uL NaOH) and four samples labeled 1, 2, 3 and 4 were prepared.  5 ug (10 uL), 10 ug (20 uL), 15 ug (30 uL), and 20 ug (40 uL) of hydroxyprolene standard samples were prepared from the hydroxyprolene stock solution with 2N sodium hydroxide for a total volume of 50 uL per sample.

2.  450 uL of Chloramine-T was added to the samples, and mixed gently.  25 minutes was allowed to elapse at room temperature for oxidation to take place.

3.  500 uL of Ehrlich's aldehyde reagent was added to each sample and mixed gently.  All samples were heated for approximately 20 min in a water bath.

4.  Three UV-VIS Spectrophotometers were used to read each sample at 550 nm.

Assay Procedure II:

Assay Procedure I does not take into account of pH change for each sample when different amounts of NaOH is added.  A change in pH could affect the rate of how hydroxyprolene oxidizes with chloramine-T and interacts with Ehrlich's aldehyde reagent.  We also would like to see if we can find a curve with a more dilute standard of hydroxyprolene.  Therefore, I developed a new procedure for finding a standard curve using 10 samples and one blank.  Since H₂O is neutral in solution, it can be changed in relation to the hydroxyprolene added without adversely affecting the overall pH.  For example, each sample will have 100 uL of NaOH, which will never change.  4M NaOH was developed for that purpose.  The total volume for each sample will be 200 uL when hydroxyprolene and water are added, and the concentration of NaOH in each sample becomes reduced to 2M.  The samples are numbered 1 through 10.  Sample 1 will have 10 uL of 0.005 g/L hydroxyprolene developed by diluting the original stock by a factor of 100.  The next sample will have an increase increment of hydroxyprolene by 10 uL.  Water is added to each solution for a total of 200 uL.  To avoid wasting the chloramine-T and Erlich's aldehyde reagent, 50 uL of each sample will be drawn into a second set of test tubes.  This allows the same volume of chloramine-T and Erlich's reagent as in Assay Procedure I to be used, instead of quadrupling the volume of these reagents to test color change in the first batch of hydroxyprolene/NaOH samples.

Results:

Ehrlich's aldehyde reagent alone is characterized by a bright yellow liquid.  There are no observable differences in each of the samples prior to heating after adding Ehrlich's reagent; each sample looks to have the same shade of color as the blank, which is a bright yellow liquid.  However, after heating the samples with Ehrlich's reagent, the color of each sample changes to a darker shade of yellow.  Sample 1 underwent the least color change, while Sample 4 underwent the most color change.  The absorbance is directly related to the amount of hydroxyprolene added and is linear based on observing the change of color shade of each sample by eye, and the absorbance detected by the UV-VIS.

Three UV-VIS instruments were used to obtain absorbance values:

A: Varian DMS100 Spectrophotometer

B: LKB Biochrom ULTROSPEC 4050

C: Genesys8 Spectronic Unicam

Assay Procedure I: The results in the order of 5ug, 10ug, 15ug, and 20ug hydroxyprolene, respectively:

A: 0.169, 0.256, 0.296, 0.404

B: 0.178, 0.270, 0.309, 0.426

C: 0.040, 0.062, 0.097, 0.105

All instruments were set at a wavelength of 550 nm.

Assay Procedure II:

HYP (uL): 0.050, 0.100, 0.150, 0.200, 0.250, 0.300, 0.350, 0.400, 0.450, 0.500

A:               0.096, 0.106, 0.120, 0.122, 0.126, 0.125, 0.129, 0.130, 0.146, 0.146

B:             -0.001, 0.002, 0.003, 0.005, 0.002, 0.002, 0.002, 0.003, 0.018, 0.010

C:              0.004, 0.008, 0.012, 0.017, 0.019, 0.020, 0.022, 0.054, 0.039, 0.042

Unknown Samples: TBA

Observations and Discussion:

Assay Procedure I: Part of the purpose of me performing this experiment was to test each UV-VIS instrument for proper functionality.  The newest instrument (C) shows a linear relationship, but does not show related absorbance values compared to instruments A or B.  Instrument B seems to always show higher absorbance values than Instrument A.  I surmise that for each instrument, the intensity of the source is different, but all instruments properly report a direct relationship of the samples.  This procedure is short in taking consideration in effects of pH change.

Assay Procedure II: No instrument for this experiment shows a perfect curve or line, but instruments A and C does show a general increase of absorption with increased addition of hydroxyprolene, which is sufficient to determine hydroxyprolene content of some unknown collagen samples.  Deviation from the expected line probably shows that the hydroxyprolene molecules were not uniform in such a dilute solution.  Instrument C shows the best linear relationship for the first 4 samples, but samples 5, 6, and 7 are "lazy" and falls below the trendline.  Sample 8 absorbance is far above the trendline, but samples 9 and 10 seems to return back to the trend shown by samples 1-4.  The least squares from instrument A show the exact same slope as the least squares from instrument C.  Also, instrument B is not sensitive enough to determine absorption values on such a dilute concentration of hydroxyprolene, so those results are bad.  All results were placed on an excel sheet, from which were were able to obtain a least squares analysis:

A: R²=0.8985; y=0.0048x + 0.0979

C: R²=0.7973; y=0.0048x - 0.0025

Unknown Samples: TBA

Suprisingly, the Varian instrument (A) shows the best prediction constructed from the modelled values because of it's R² value closer to 1 than the results obtained by the newer Genesis8 (C) instrument.  However, this value does not account for the fact that the Genesis8 is better at not fluctuating between numbers when making measurements than the Varian.  The Varian also tended to deviate from zero with the blank solution.  Overall, I was pleased with the overall results of both instruments.

For further discussion, please refer to the full article published in Clinical Biochemistry, Volume 29, June 1996