The ability to measure single cell growth is of fundamental importance to our understanding of biological systems. Despite several major efforts, studying this phenomenon has remained largely intangible due to the simple fact that cells are small and only double their mass during their lifetime. In order to draw conclusions on growth trends and gain insight on growth regulatory systems, the required sensitivity to mass is in the order of femtograms.

Introduction

The question of how single cells regulate their growth has been described as "one of the last big unsolved problems in cell biology" [1]. The major question has been whether cells grow at a constant (linear) rate or proportional to their size (exponential). This question is essential to understating how a constant cell size distribution (homeostasis) is maintained. If the growth is exponential, any small difference in the size of two daughter cells at the time of division would be compounded with each generation. Thus, a growth trend which is proportional to the cell size requires a robust, intrinsic regulatory system to maintain homeostasis. On the other hand, if the growth is constant no such special mechanism is necessary. It has been calculated that measuring the difference between these two trends requires sensitivities of less than 6% of the cells size. For single cells this translates to a required sensitivity on the order of femtograms.

Until recently, the only reliable method to address this problem has been to measure cell size distributions using an impedance counter and apply statistical analysis to draw conclusions on growth rates [2]. Such approaches assume that measuring volume is analogous to measuring mass and furthermore limit understanding of the inherent cell to cell variability present in any population. Over the last two years several new methods that utilize Microelectromechanical systems (MEMS) technology have been introduced which are capable of measuring single cell mass with the required accuracy [3, 4]. However, these methods have their own limitations: they are either capable of measuring non-adherent cells with high throughput but no lifecycle information at the single cell level, or measuring adherent cells, but one at a time.

In light of these efforts, it is clear that the ideal method for measuring cell growth should have the ability to follow single cells throughout their life-cycle, be compatible with many cell types (adherent/non-adherent, prokaryotes, eukaryotes, etc.), be non-invasive and also work on spatial scales ranging from the population level (millimeters) down to subcellular (microns). Here we show that Spatial Light Interference Microscopy approaches this ideal. SLIM can be used to quantitatively measure dry mass with femotogram accuracy for a variety of cell types, over several cell cycles and at spatial scales ranging from millimeters to sub-micron.

Optical Measurement of Cell Mass

In the 1950's, soon after the invention of phase contrast microscopy, it was shown that the refractive index of a solution is linearly related to the concentration of the solute by a quantity known as the refractive increment [5]. Furthermore, it was also shown that for biological samples this quantity varies insignificantly, such that it is reasonable to consider one average value for all cellular contents. Following this demonstration, which was largely due to experiments in immersion refractometry, it was realized that if one measured the optical path length difference produced by an object using interference microscopy, it is possible to calculate the objects dry mass per unit area [5]. For the case of living cells, which are heterogeneous objects, the dry mass per unit area may then simply be integrated over the entire projected area of the cell.

Although the idea of using interference microscopy to measure cell mas is old, the applications have been limited due to two major reasons. Firstly, the spatial resolution and contrast are limited due to the speckle generated by coherent laser sources. Secondly, the experimental setups required for interference microscopy are traditionally very complicated and are too unstable for long term biological imaging. In recent years these limitations have been overcome through the advent of common path interferometry, to address the issue of stability and broadband phase measurements, to address the issue of speckle [6].

Spatial Light Interference Microscopy (SLIM)

SLIM is a new technique that is both common path and broadband [7]. SLIM combines two classical ideas in optical imaging, Zernike's Phase Contrast microscopy, which provides high contrast images of phase objects and Gabor's holography which gives quantitative phase information about the objects. In short SLIM operates by measuring the intensity of the image at four different phase shifts between light scattered by the object and the un-scattered light (fig. 2). From the four intensity measurements a unique quantitative phase map may be determined [7]. Since SLIM uses white light (short coherence length) illumination it provides speckle free imaging. Due to the illumination and common path geometry, the temporal and spatial noise in optical path length has been measured to be 0.029 nm and 0.28 nm respectively. When translated to dry mass the spatial and temporal sensitivities are 1.5 and 0.15 femtograms/µm2 respectively. Furthermore, since SLIM is designed to be an add-on module to a commercial microscope (fig. 2) it is possible to utilize other commonly used modalities such as fluorescence, which is essential for biological studies. The high demand for long term biological imaging has also pushed several companies to develop microscope top incubation systems which enable SLIM to measure living biological samples for weeks at rates of several Hertz, providing a large temporal range. By scanning and stitching it is possible to image an entire microscope slide with subcellular resolution, thus providing information on both the growth of the entire cell population and single cells simultaneously. Due to its sensitivity, multi-modal capabilities, flexibility in spatial and temporal scales, and non-invasive nature, SLIM approaches the ideals for measuring cell growth that were laid out earlier.