Pontier and Schweisguth 2015.

In response to “A Wolbachia-Sensitive Communication between Male and Female Pupae Controls Gamete Compatibility in Drosophila”

Caveat:This paper is one of the most incomprehensible papers I have ever read. But because a reviewer cited this in contrast to my work I am forced to read it and point out its flaws.

The Hypothesis
This article proposes that the basis for Cytoplasmic Incompatibility lies in the idea that Wolbachia can disrupt pheromone signaling of insect oenocytes, (epidermal secretory cells) between sexes, during pupation; and this causes downstream effects in testes, altering germline stem cell replication rates, causing an undefined “gamete incompatibility.” Yes, the hypothesis is that convoluted… Regardless of the data, the hypothesis is obviously wrong because it contradicts not only substantial literature, but also basic insect physiology and development on multiple points.

1) Oenocytes (pheromone secretory cells) derived from ectodermal developmental tissues are a part of the epidermis.1 This immediately raises the question, of “How does Wolbachia, which infect internal reproductive tissues, influence mechanics of epidermal cells? This hole adds a layer of complexity which the authors are unwilling to attempt to rationalize or even address.

2) Oenocytes are vastly different in shape, placement, and complexity between insects.1 Furthermore, pheromone sex signaling between insects is also extremely unique, complex, and can differ vastly even between subspecies and intra species biotypes. Insects use unique pheromones to attract species specific mates! The fact that Wolbachia from one insect can be artificially established in another cross genera/family insect and still cause CI raises the question, “How can heterologous Wolbachia infections induce CI by pheromone signaling when the cellular structures and chemicals used are NOT conserved among insects and are in fact completely unique?”

3) The paper rewrites pheromone chemical ecology of insects by being the first paper in history to demonstrate sex specific pheromone communication between insect pupae. The authors say this proudly, “To our knowledge, [this paper is] the first indication that a between-sex communication occurs prior to adulthood.” Apparently, everyone in history has missed this conserved pathway in insects; In reality this makes the data unbelievable because it would have earth-shattering effects in chemical ecology.

4) CI based upon pheromone signaling would imply a distance limit on CI induction at which no CI would occur if pupae were reared far enough away from the chemicals. This is obviously false.

5) They never define what “Gamete incompatibility” means to them. Assuming they are right, how does a pheromone, which according to their hypothesis only effects the rate of stem cell mitosis, cause the cytology observed in CI? How does lowering the speed of replication of germline stem cells cause them to become incompatible? I point out that this observation of lowered germline stem cell replication is in direct contrast to Fast et al., Science 2011 (Frydman’s Group) which showed that Wolbachia actually increase germline stem cell replication rates. How do the authors justify the hypothesis in the context of this seemingly opposite publication?

5) No cytological data linking to canonical CI cytology. Period. You can no longer claim a CI connection without this. Standards have changed…

6) Their own data disproves their own hypothesis. Take a close look at Figure 2B. Their argument is that males pupating in the presence of the phantom chemical secreted by infected female pupae (Blue filled male) have lowered stem cell ratios. Immediately next to this data they present data from uninfected males in the presence of ZERO female pupae (White male) which has the EXACT same stem cell ratio. Thus treatments with presence of the phantom chemical or absence of the phantom chemical are exactly the same disproving their hypothesis.

7). Statistical significance does not equate to a hypothesis being right. The entire hypothesis is based upon countless minute observations with very little real difference between them.  Conveniently, in the most important measurements, that variance gets massive.