Michael Welte

Associate Professor

Contact Information:

University of Rochester
Department of Biology
River Campus Box 270211
Rochester, New York 14627-0211

Hutchison 317 (office)
Hutchison 311 (lab)

(585) 276-3897 (office)
(585) 276-2440 (lab)

Michael Welte

Research Overview

Our laboratory studies how microtubule motors, such as kinesins and dyneins, transport organelles and other cargoes throughout the cell. We focus on elucidating the mechanisms by which cells control specificity, timing and destination of transport. figure 1
Lipid droplets (yellow) in early Drosophila embryos. Droplet distribution is controlled by active transport along microtubules. To address these general questions, we employ a powerful new model system that can be studied with a battery of techniques, from genetics to biochemistry to biophysics: the motion of lipid droplets in Drosophila embryos. We employ this system to discover regulators that control transport and to determine their molecular mechanisms of action. Many of the molecules involved in droplet transport are also important for other processes, and we employ the insights from our analysis of droplet transport to dissect the motion of selected other cargoes.

Lipid droplets store neutral lipids when resources are abundant; when resources are scares, stored lipids can be used for the production of energy and as a source of metabolites during “lean” times. As a consequence, our studies illuminate not only how microtubule motors are deployed in a controlled manner in the cell, but may also shed light on lipid homeostasis and energy metabolism. We recently discovered a new biological role for lipid droplets as sites of regulated protein sequestration, a role with broad implications for cell and developmental biology.

Our research is very visual and generates striking images; see, for example, the cover of the September 30th (2003) issue of Current Biology and our entry into the Drosophila Image Award 2007 competition.

For more details on our research, check out our lab website.

Regulation of lipid-droplet motion


figure 2
Molecules involved in droplet transport. Motors: kinesin-1, cytoplasmic dynein. Coordinators: Klar, dynactin. Conductors: LSD2, Sfo. Signals: Halo; uncharacterized signal dependent on Ago2.
In the early Drosophila embryo, lipid droplets move bidirectionally along microtubules, powered by plus- and minus-end motors. In collaboration with our colleagues, we have identified molecules that appear to control transport at three distinct levels (see cartoon): “coordinators” (pink) ensure that at any given moment only motors for one direction are active; “conductors” (purple) determine how frequently this machinery switches between the states in which plus- or minus-end motors are active; “signals” (orange) modulate this switching frequency in trans. Since droplets are moved by the widely employed motors kinesin-1 and cytoplasmic dynein, mechanistic insights into this regulatory machinery will likely illuminate transport regulation in general.

We are now determining how these regulators act at the molecular level. Ongoing projects include

Transport regulation in oogenesis     figure 3
The coordinator Klar (green) is present on the nuclear envelope in ovaries (DNA in blue). Guo et al.,2005


Several of the regulators involved in droplet transport are also expressed during oogenesis. Our genetic analysis indicates that here they serve roles in a number of processes, from trafficking of secretory vesicles to the positioning of the oocyte nucleus. We focus on the function of Klar: Oocytes express four different Klar isoforms, and our analysis indicates that these isoforms have distinct biological properties. We are now testing whether different isoforms are dedicated to distinct transport processes or uniquely regulate a common set of transport processes or both. The long-term goal of this analysis is to understand how cells can adapt a small set of motors and regulators to uniquely control a wide range of transport processes.

Lipid droplets as protein sequestration sites


We recently found strong evidence that lipid droplets in early Drosophila embryos figure 4
When living embryos are centrifuged, it is possible to separate organelles by density in vivo. This technique reveals that certain histones are sequestered on lipid droplets. See Cermelli et al., 2006, for details. serve as storage depots for maternally provided histones (Cermelli et al., 2006). Histones are massively present on lipid droplets in ovaries and early embryos, but not in late embryos or cultured cells. Quantification of histone levels and t ransplantation experiments show that these histones are not irreversibly trapped, but can be transferred from droplets to nuclei as embryogenesis proceeds. We are now testing how widespread protein sequestration on droplets is, what functions it might serve and how specific proteins are recruited to lipid droplets in a reversible manner (see Welte, 2007, for discussion).

Selected Publications