Di Carlo Microfluidics
UCLA Di Carlo

We are exploiting unique physics, microenvironment control, and the potential for automation associated with miniaturized systems for applications in basic biology, medical diagnostics, and cellular engineering.

1. Inertial Microfluidics

Di Carlo Inertial Microfluidics

In microfluidic systems, inertial, nonlinear terms of the Navier-Stokes equations describing fluid flow are often neglected. This is because of the widely-held notion that because of the small length scales the Reynolds number in these systems is concurrently small. This is not necessarily the case, especially when microfluidic systems are operated at high flow velocities. In these finite inertia systems, particles in a fluid exhibit several unique behaviors.  Under laminar flow particles migrate across streamlines in a predictable and precise manner. Ordering of particles into regular trains along the longitudinal direction of the channel can also be observed in these systems. The physical basis of focusing and ordering phenomena in inertial flows has been explored briefly, but an intuitive understanding is still elusive. Additional secondary flows induced by particles in these finite inertia confined flows are also of interest.  Further understanding and intuition for systems would allow application to a broad base of engineering and biological problems. A particularly good match for this technology is in developing novel flow cytometry systems with massive throughput, enabled by ordering and inertial focusing without added sheath fluids and the possibility of automated intrinsic solution exchange and cell sample preparation. These systems should be useful in rare cell detection from clinical samples, with principle uses in cancer diagnostics. Inertial microfluidics for concentration of rare cells or particles from large volumes of bodily fluids is also promising given the high-throughput and unique separation forces that increase with throughput. 

Funding (NSF, DARPA, DoD, Coulter Foundation), Collaborators (Aydogan Ozcan, UCLA; Bahram Jalali, UCLA; Edward R.B. McCabe, UC; Eric P.Y. Chiou, UCLA; Howard A. Stone, Princeton; Steve Graves, UNM)

References: Hur, Tse, and Di Carlo. Lab on a Chip 2010; 10:274-280. Di Carlo. Lab on a Chip 2009; 9:3038-3046.  Gossett and Di Carlo. Analytical Chemistry 2009; 81:8459-8465. Russom et al. New Journal of Physics 2009; 11: 075025. Di Carlo et al. Physical Review Letters 2009; 102:094503.  Edd et al. Lab on a Chip 2008; 8:1262-4.  Di Carlo et al. Proceedings of the National Academy of Sciences USA 2007;104:18892-1889.

Dino Di Carlo


2. The Mechanics of Cancer and Metastasis

Metastasis of cancer to secondary sites, not the primary tumor is responsible for 90% of cancer-related deaths, representing over 500,000 deaths per year in the United States alone. Consequently, metastases are among the most important biological problems to address in cancer research. Although a link between the primary tumor and secondary sites has been discovered in the form of cancer cells circulating in the blood (CTCs – circulating tumor cells), little is known about the mechanism by which cancer cells pass through the barrier of endothelial cells and extracellular matrix lining vessel walls, migrate through the confined environment, and detach to circulate freely in the peripheral blood. An understanding of this process of intravasation could lead to novel therapies targeted at the initial stages leading to metastases. However, direct observation of the process has been mostly unsuccessful because of the technical difficulty of visualization in a 3-D living organism. Microfabricated in-vitro systems that can replicate particular aspects of the tumor environment quantitatively and allow observation of the process of intravasation can yield critical insights into this problem.

Di Carlo Cell Mechanics Cancer Mechanical Mutagenesis
Figure - Cell Division in Mechanically Confined Microenvironments.  (A) Schematic representation of the tumor microenvironment (unchecked cell division leads to high internal pressure and stiffness.  (B-C) Experimental methods to investigate the effect of the mechanical microenvironment on cell division success and symmetry.

Additionally, failure in cell division leading to improper chromosome segregation can lead to aneuploidy, chromosomal instability and progression to malignancy.  There remains the question of what causes these particular failure modes in cell division in vivo, and whether the unique tumor microenvironment contributes to the likelihood of these events.  Notably, cells in tumors are exposed to increased interstitial pressures and tissue stiffness as cells grow unchecked by contact inhibition (Figure above).  Given that the processes of chromosomal segregation and cytokinesis are inherently mechanical, we are investigating the effects of the mechanical microenvironment on these processes.  We are evaluating whether the unique mechanical microenvironment in tumors ultimately contributes to genetic instability, and would suggest that treatments affecting the tumor microenvironment may assist in combating early phases of tumor progression

Di Carlo Westbrook Weaver Tse Henry Tat Kwong Cell Mechanics Mutagenesis


3. Automated Cell Biology

Ultimately, our understanding of biological systems and our ability to apply this understanding to improve human health or quality of life is limited by our tools to explore and report about the micro & nano world at which life functions. We believe that a transformative rather than incremental change to biomedicine will require a critical mass of reliable quantitative data describing the dynamics of cellular processes. Being able to answer questions such as (How much of protein X is present in a particular cell? Where is biomolecule Y located in space and time within the cell? What happens if more of protein X is added near focal adhesions?) is only possible for a small subset of proteins, and only in isolation from other perturbations. Obviously, it is a daunting challenge to explore the large number of permutations of multiple biomolecules in space and time, and a dedicated effort with sustained resources (perhaps similar to a human genome project) will be required to achieve this. As we are able to more precisely answer questions such as those above for arbitrary biomolecules, we are confident that this will impact a variety of disciplines (see figure below), with the premise that a spectrum of new tools will be necessary to address this challenge

A large step in achieving a quantitative understanding of biological function can be gained by understanding the effect of localization of biomolecules or mechanical stimuli on cell behavior. This is an area where automated platform tools with significant throughput are lacking, but is of critical importance given the re-use of signaling pathways and second messengers in signal transduction across the cell.

Di Carlo Tseng Automated Biology

Figure - Quantitative Tools in Biology. Past tools such as DNA sequencers, flow cytometers, and microarrays have allowed improved quantitative understanding of biological systems. A variety of new high-throughput tools will be necessary in the coming decades (example break-through instruments include high-throughput binding energy measurements* Maerkl and Quake Science 2007, and our work for micro to nanoscale cellular perturbation). These new tools are poised to allow breakthroughs in fields from biofuel production, gene therapy, to cell-computer interfaces.

We are developing a generally applicable technique with the ability to selectively and dynamically modify the local environment within single cells with nanoscale precision for hundreds of thousands of cells in parallel to obtain statistically useful data.  This platform would provide active spatial control of chemical and mechanical signals for arbitrary proteins with real-time dynamic exploratory experimentation that is poised to provide transformative quantitative data.  

Funding (NIH New Innovator Award), Collaborators (Jack Judy, UCLA)
References: Tseng et al. Nano Letters 2009; 9: 3053–3059

4. Microfluidic Directed Cellular Evolution

Directed evolution of improved proteins and enzymes for applications in biotechnology, food science, and medicine has been very successful over the last few decades. Selection processes have been implemented that select for thermostability, substrate specificity, kinetics, and other properties in evolved enzymes. In most cases evolutionary processes have been applied to optimize a particular protein’s characteristics but not a bulk cellular behavior that has contributions from a system of interacting proteins. This may be partly because of a lack of suitable selection techniques for particular cellular behaviors. Microfluidic technologies may offer advantages in creating new useful selection criteria for cellular evolution. Examples include cell migration speed, proteolytic activity, deformability, shear-stress stability, and osmotic tolerance. Besides gaining an understanding of dominant molecular pathways in controlling these behaviors, the resultant evolved cell populations and genetic modifications may be useful for therapeutic applications. For example, fast migrating and proteolytically active cells and the underlying genetic circuits may be used in aiding and speeding wound healing, as cell migration into the wounded site can be a rate limiting step for further healing. Additionally, directed evolution in prokaryotic cells may assist in the understanding of chemotaxis, adherence, and sporulation and be therapeutically useful in designing pharmaceuticals, antisense, or gene therapies to disrupt these important behaviors. 

Collaborators (Vladana Milisevljevic, Jeff Miller, UCLA)