Harold Paul Erickson
James B. Duke Professor of Cell Biology

Cytoskeleton: It is now clear that the actin and microtubule cytoskeleton originated in bacteria. Our major research is on FtsZ, the bacterial tubulin homolog, which assembles into a contractile ring that divides the bacterium. We have studied FtsZ assembly in vitro, and found that it assembles into thin protofilaments (pfs). Dozens of these pfs are further clustered to form the contractile Z-ring in vivo. Some important discoveries in the last ten years include:

•    Reconstitution of Z rings in vitro. We provided FtsZ with a membrane tether, and found that when incorporated inside liposomes, FtsZ-mts can assemble Z rings without any other proteins.

•    These reconstituted Z rings generate a constriction force on the membranes, again without any other proteins (no motor molecules).

•    The constriction force is generated by a curved conformation of FtsZ pfs generating a bending force on the membrane.

Important questions for the future are:

  • How are FtsZ pfs arranged in the Z ring? We favor the ribbon model, where pfs are parallel and laterally associated into a ribbon. Many others in the field favor a scattered model, where pfs are more widely separated. We are exploring new electron microscopy (EM) methods to resolve the structure. We have also developed new tools to facilitate superresolution light microscopy (PALM).
  • How does FtsZ treadmilling work? Our lab provided the first evidence that FtsZ treadmills, adding subunits at one end and losing them at the other (Redick J Bact 2005). This has now been confirmed in vitro and in vivo. We are developing theoretical models and experimental (EM) methods to determine the detailed mechanism of treadmilling.
  • What is the structure of the septum in dividing bacteria? There is wide agreement that Gram-positive bacteria divide by ingression of a plate-like septum. Conventional EM suggests that Gram-negative bacteria have a shallower V-shaped constriction. We are revisiting this using novel fixatives and high-pressure freezing for thin section EM.
Irisin. We believe the irisin story is bunk. Irisin was proposed in 2012 as a novel myokine, secreted by muscle cells in response to exercise, it induces the transformation of white fat to brown fat. This inspired hopes of an exercise pill that might correct obesity and other metabolic disorders. We have argued that the original discovery was flawed in several respects (Erickson, Adipocyte, 2013), and that the 300+ published follow-up studies are based on flawed commercial antibodies (Albrecht et al, Sci Rep 2015). We are now developing new assays to determine if irisin exists in in the blood of humans, primates and other animals. We expect it does not, especially in humans whose FNDC5 gene has a mutated start codon.

Current Research Interests

Updated June 2019. 

Cytoskeleton: It is now clear that the actin and microtubule cytoskeleton originated in bacteria. Our major research is on FtsZ, the bacterial tubulin homolog, which assembles into a contractile ring that divides the bacterium. We have studied FtsZ assembly in vitro, and found that it assembles into thin protofilaments (pfs). Dozens of these pfs are further clustered to form the contractile Z-ring in vivo. Recent work from other labs has established that the pfs are clustered in small patches that move around the circumference of the cell by a treadmilling mechanism. A major project in our lab is to generate a Monte Carlo model that can explain treadmilling and nucleation, based on conformational changes and GTP hydrolysis. (See my 2019 article on microtubule assembly from flared pfs for a preview of our treadmilling ideas, and an important implication for microtubule assembly.) We are also interested in how FtsZ generates a constriction force, and how this force constricts against turgor pressure, or how it “cheats turgor.”

Irisin. We got interested in irisin from an indirect pathway. We believe the irisin story is bunk. Irisin was proposed in 2012 as a novel myokine: secreted by muscle cells in response to exercise, it induces the transformation of white fat to brown fat. This inspired hopes of an exercise pill that might correct obesity and other metabolic disorders. We have argued that the original discovery was flawed in several respects (Erickson, Adipocyte, 2013), and that the 600+ published follow-up studies are based on flawed commercial antibodies (Albrecht et al, Sci Rep 2015). We are currently using mass spectrometry to measure plasma concentrations of irisin in humans and baboons. This  should be interesting because humans have a mutation in the start codon that should reduce its expression ~100x relative to other animals.

Current Appointments & Affiliations

Contact Information

Some information on this profile has been compiled automatically from Duke databases and external sources. (Our About page explains how this works.) If you see a problem with the information, please write to Scholars@Duke and let us know. We will reply promptly.