Lineage dynamics of the endosymbiotic cell type in the soft coral Xenia

Corals form an endo-cellular symbiotic relationship with Symbiodiniaceae. What’s the cell type involved in the symbiotic relationship? How does this symbiosis happen? In our recent published work, with single cell RNA-seq approach, we start to shed lights on these fundamental questions.
Published in Ecology & Evolution
Lineage dynamics of the endosymbiotic cell type in the soft coral Xenia

Share this post

Choose a social network to share with, or copy the shortened URL to share elsewhere

This is a representation of how your post may appear on social media. The actual post will vary between social networks

This project was initiated about three years ago by two non-marine biologists, Chen-Ming Fan (Fan) and Yixian Zheng, who hoped to develop model organisms to study coral-algae symbiosis at molecular and mechanistic levels. At around this time, I arrived in Yixian’s lab as a new postdoctoral fellow. With a Ph.D. in development biology and a bachelor’s degree in bioinformatics from Zhejiang University, China, my goal was to develop an exciting basic research program in developmental biology so I can launch an independent research career. As I was trying various projects involving mammalian model, Yixian suggested to me to consider building a model organism from scratch to study coral-algae symbiosis. Since I was interested in applying my bioinformatic training in my research, I jumped on this opportunity. By trying to grow about 9 different soft and hard corals in our lab aquarium over half a year,

Illustration of lab aquarium. Photo credit, Jeremy Hayes

it appeared that one of the soft coral Xenia grew the fastest and was easy to manipulate. So I decided to work with Xenia.Since I have never assembled any genome and transcriptome from scratch, Xiaobin Zheng, a highly experienced bioinformatician and mathematician, stepped in to aid my analyses.

Illustration of Xenia colony. Photo credit, Ed Hirschmugl

As I am familiar with DNA extraction from zebrafish and mouse, I thought it would be very easy to isolate Xenia genomic DNA and assemble its genome. But I had a tough beginning in 2017 because I wasn't able to obtain good quality and quantity Xenia DNA. At that time, Fan brought up a point I had never thought about, the various chemicals, such as Ca2+, in the seawater could facilitate DNA degradation during isolation. Fan was exactly right, after washing Xenia with Ca2+ free sea water, I was able to obtain high quality and quantity Xenia DNA! I also noticed that sea salt used to make sea water for corals also affected other procedures, so washing Xenia became my first step for trouble shooting.

My fortune started to change for better in the spring of 2018 when a new postdoc, John Urban, joined our department. John had a lot of experience with Nanopore sequencing and suggested to use DNAzol to isolate long DNAs and several genome assembly methods. Coupling these with Hi-C, I was able to assemble a highly contiguous Xenia genome.

At the time I was finishing the Xenia genome assembly, our department purchased a 10x single cell (sc) RNA-seq platform which provided me huge convenience to perform scRNA-seq in Xenia. I tried different enzyme combinations to dissociate Xenia polyps into single cells, but was met with low dissociation efficiency. Xenia secretes a lot of mucous, especially upon manipulation. I suspected that the mucous could inhibit the dissociation reactions. At that time, a new postdoc, Aki Ohdera, joined our lab. He is a marine biologist studying another cnidarian, Cassiopea. Aki suggested to add 4% cysteine to neutralize the mucous in my cell dissociation reaction. This greatly increased the digestion efficiency and reduced digesting time from 4 hours to less than 1 hour! With this, I was able to perform Xenia scRNA-seq. Having good colleagues was my fortune, and I believe I was also lucky to have chosen Xenia coral. Recently, we found that the10x platform did not work well for scRNA-seq for hard corals and Cassiopea. Clearly different scRNA-seq strategies will be needed for different cnidarians.

A major challenge in applying fluorescence microscopy to study symbiotic organisms like corals is the autofluorescence emitted by the endosymbiotic algae living inside the coral and the fluorescent proteins expressed by the corals. In the course of developing protocols for antibody, BrdU, and EdU staining and imaging, I found that the autofluorescence specifically emitted by the algae in the far-red channel was extremely strong. My training in developmental biology allowed me to turn the deficiency for fluorescence microscopy to an advantage for fluorescence activated cell sorting (FACS). Based on the specific alga fluorescence, I was able to isolation the alga-carrying Xenia cells by FACS and identify their transcriptome, which enabled the characterization of the endosymbiotic Xenia cell type.

Corals such as Xenia are very good at regeneration. Before my arrival, Yixian played with Xenia regeneration and noted that stalk fragments regenerated faster than tentacles. I found that adherence of coral fragments to surfaces facilitated regeneration and the variability in attachment greatly influenced the consistency of regeneration speed and efficiency. To overcome the variability, I transplanted the polyp into 24 well plate and allowed it to attach to the plate surface before cutting away all the tentacles. This simple step has greatly helped me to take advantage of regeneration to study how the endosymbiotic Xenia cells progress from progenitor cells to algae containing cells and eventually lose their algae. 

For Xenia to be a good model to study symbiosis, it is critical to develop methods to manipulate genes. I have made some inroads in reducing gene expression in Xenia using RNAi, a method developed in my host department over a decade ago. Please contact me if you are interested in this. I am actively working on establishing Xenia cell lines for gene manipulation. Refining scRNA-seq to better characterize the 16 cell clusters we defined in our paper and identifying additional rare Xenia cell types are also on my to-do list.

I am also particularly interested in identifying calcification-related cell types in corals in general and hope to perform scRNA-seq on reef building corals in the field to identify gene signatures that characterize bleaching and post bleaching recovery. I hope this blog platform will help me to establish collaborations with field coral biologists for this effort.

Carnegie Institution’s Embryology Department has been home for model organism development for decades. I appreciate the amazingly interactive and helpful environment in the department. I want to take this opportunity to thank Allison Pinder for running an efficient sequencing core, Mahmud Siddiqi for supporting all microscopy and FACS, Mike Sepanki for electron microscopy, and Fred Tan for helping with all the coding and setting up the Carnegie Marine Organism Portal ( For detailed studies, please refer to: Lineage dynamics of the endosymbiotic cell type in the soft coral Xenia (link:

Please sign in or register for FREE

If you are a registered user on Research Communities by Springer Nature, please sign in

Go to the profile of Tegan Armarego-Marriott
almost 4 years ago

This is such a great post, thanks Minjie!

Go to the profile of Lily Wong
almost 4 years ago

Hi Minjie! I came across your work at the recent Society for Developmental Biology annual meeting. I enjoyed your talk very much! This blog post was very illuminating on the steps you took along the way to bring this project to its current state and beyond. I work with planarians which also secrete a lot of mucus. There are protocols developed for immunohistochemistry, RNAi, ISH in this model system, please check them out. I feel inspired by the work you and your colleagues are doing.

Go to the profile of Minjie Hu
almost 4 years ago

Hi Lily, thanks for the information. I will definitely look into it. I noticed you didn't mention IF in planarian. Is it also hard to do due to the high autofluorescence?

Go to the profile of Lily Wong
almost 4 years ago

Immunofluorescence is done in both whole mount and tissue sections in planarians. It takes time to optimize for each antibody because one needs to experiment with 1) the chemical used for mucus removal, and 2) the type of fixative used, and 3) the blocking agents for immunolabeling.  You can consult this protocol book for details: Planarian Regeneration, Methods and Protocols, Editor: Rink, Jochen