The use of principal component analysis and self-organizing map to monitor inhibition of calcium oxalate crystal growth by Orthosiphon stamineus extract

Abstract

The modified Schneider's gel slide method is used to monitor the inhibitory effect of 50% methanol extract of Orthosiphon stamineus on the growth of calcium oxalate crystals. The images of crystals grown in the gel were captured at 1, 2, 3, 4 and 24 h with digital camera and image analysis was carried out. A total of nine variables relating to size and shape parameters were calculated and six were used for further analysis. Principal component analysis (PCA) and self-organizing map (SOM) were applied in the visualization of the size and shape distribution of the produced crystals using the following six variables: area, convex area, aspect ratio, equivalent diameter, roundness and full ratio. The results indicate that the modified gel slide method with the use of PCA and SOM to reduce the dimensionality of the data allowed an intuitive presentation of the differences in the studied inhibitors. The decrease in crystal growth for the extract and positive control of sodium citrate was clearly evident in the principal component scores and also on the location in the SOM. Evidence about the decrease in growth was also provided by an inhibition index of 0.290 for the extract and 0.255 for the positive control.

Introduction

There are many theories for the formation of kidney stones. One of them, the crystallization precipitation theory, implies that supersaturation of urine leads to precipitation of stone crystallites. These critical particles become entrapped and subsequent crystal growth follows. The process of aggregation and secondary nucleation leads to the formation of large crystals, which ultimately develop into kidney stones [1]. Another theory, the inhibitory theory, suggests that normal urine contains substances that inhibit the crystallization of calcium oxalate. Inhibitors of crystal growth block the growth of crystals and prevent stone formation [2].

Other than normal drugs, herbals are also commonly used to treat urinary stones and studies of such local Malaysian medicinal plants confirm the presence of compounds that show inhibitory effect in growth of calcium oxalate [3].

Among the methods to measure the inhibitory activity of crystals are photometry [4], turbidimetry [5], mixed suspension mixed product removal MSMPR [6], use of Coulter counter [7], [8], [9] and spectrophotometry [10]. Each of this method has some advantages but there is no ideal technique. A comparison of seven different methods to measure crystallization in urine were reported in a workshop in 1987 and the approach of crystallization in gels contained in wells of microtitre plate gave good efficiency and was suggested for basic research and clinical routine [11]. In a subsequent workshop [12], problems relating to crystallization measurements were discussed, for example, Coulter counter analysis was laborious, and optical techniques involving densitometry approaches required high crystal density. However, most of above techniques measured inhibition indirectly and not the effect on single crystals.

Nowadays, the use of image analysis offers a way to quantify the variations in crystal population. Shape and size can be characterized subsequent to the visualization of the crystals by light microscopy. The quantitative description of the morphology of the produced crystals is with the use of different shape and size descriptors. A simple method that gives a rapid insight into the variations of these descriptors is by using principal component analysis (PCA) to summarize all the shape and size parameters as carried out by Bernard-Michel et al. [13]. Another approach is the use of self-organizing map (SOM), which was reported by Laitinen and coworkers [14] as giving a better perception of the distribution compared to PCA.

In this study, a modified Schneider's gel slide method [4] was used to study the inhibition of calcium oxalate crystal growth by 50% methanol extract of leaf of Orthosiphon stamineus at concentration of 5000 ppm. A positive control of sodium citrate 10 ppm solution was also used to monitor the inhibition. A large set of data consisting of size and shape parameters was produced by image analysis while PCA and SOM were used to analyze the data. The objective of this work, other than to monitor the inhibitory effect of the plant extract on crystal growth, was also to study the feasibility of using PCA and SOM on image analysis data of modified gel slide method to monitor the effects on crystals.