Nanog Heterogeneity: Tilting at Windmills?

Mouse     embryonic  stem   cells   (ESCs) closely  approximate  pluripotent embryo founder cells  resident in the  blastocyst. However, it is important to keep  in mind that ESC propagation is a cell culture phe-  nomenon. ESCs  may  be  liberated from constraints   imposed  by   the   developmental  program in vivo, but they are also subject to stimuli and conditions that  do not  occur  in  the embryo. Depending on the specific culture  setting, ESCs exhibit different  morphology,  gene  expression, epigenetic features, and self-renewal effi- ciency  (Wray et al., 2010). Notably,  ESCs on a feeder layer present as homogenous clusters  of  small,  tightly  packed  cells, whereas without feeders and  in the pres- ence of  leukemia   inhibitory  factor  (LIF), ESCs  are flattened and exhibit heteroge- neous morphologies. A suite of transcrip- tion  factors  is  expressed  in  a  mosaic fashion  in such  feeder-free serum and LIF cultures (Marks et al., 2012). Nonetheless almost all cells express the essential plu- ripotency determinants Oct4  and  Sox2, and at the population level ESCs cultured in serum and  LIF  can  reliably  form  chi- meras and give germline  transmission.

Heterogeneity   in  transcription   factor expression is commonly observed by im- munostaining and  thus   reflects  protein  levels.  In some cases, knockin  of a fluo- rescent reporter (FP) has  been  used to infer transcriptional  regulation. Nanog  is studied  frequently  

because it plays  key roles   in  establishment   of   pluripotency, self-renewal, and reprogramming. Nanog reporters are expressed heterogeneously in ESCs cultured in serum and LIF without feeders. Furthermore, they indicate that a fraction of cells can revert from Nanog low to  Nanog  high  states (Chambers  et  al.,

2007). Similar observations for Rex1 and Stella reporters have led to the proposition that ESCs experience dynamic heteroge- neity and  that  such  metastability may be an   essential  component   of  pluripotent identity  (Hayashi  et  al.,  2008;  Toyooka et    al.,    2008). However, if inductive signaling through the fibroblast  growth factor/mitogen-activated   protein   kinase pathway   is    blocked   and    activity    of glycogen synthase  kinase 3  is  inhibited with two  small  molecules  (2i), ESCs  are highly  homogenous  yet  fully pluripotent even   in  the  absence of  feeders (Wray et  al.,  2010).  Heterogeneity and  fluctua- tion are therefore culture-induced pertur-  bations and  their relevance to potency or fate choice is questionable. Nonetheless, these phenomena continue to attract in- terest. To add  fuel to  this  debate,  it has  recently  been suggested that monoallelic expression may underlie Nanog heteroge- neity (Miyanari and  Torres-Padilla, 2012). This inference is based primarily on local- ization  of  nascent  transcription sites   by RNA FISH, although the authors also claim that it is reflected in the alternating expres- sion of fluorescent reporters.

Contrary  to these previous findings,  in this   issue    Faddah  et   al.   (2013)   now describe  a  failure to detect significant heterogeneous expression using   new Nanog:FP knockin  reporters  and  single- molecule  mRNA   FISH.   These  authors ascribe  previous results  to  artifacts  of endogenous gene disruption.   Indeed, the  authors show  some  differences be- tween  reporters—a  useful  reminder that a knockin  cannot blithely be assumed to recapitulate all aspects of normal  regula- tion. Remarkably, however, Faddah et al. did not examine ESCs  without feeders in serum  and   LIF,  and   therefore  cannot draw conclusions pertinent to the circum-  stance in  which  heterogeneity has  been documented.  It  would   be   intriguing   if their  reporter  remained  homogeneously expressed in these conditions,   unlike Nanog  protein. In a  second  report, Fili- pczyk   et   al.   (2013)   create  functional Nanog-FP  fusion proteins and  generate reporters that  are  anticipated to  mirror normal  Nanog  protein distribution. These authors  do   employ   feeder-free  culture and observe heterogeneity in serum and LIF compared to relative homogeneity in

2i. The interesting feature of this report  is that  in both  conditions  they  find  a  high correlation between reporters expressed from either allele, as also seen by Faddah et  al. (2013). This finding  therefore chal- lenges the  idea  that there  is  significant monoallelic   expression  of   Nanog    and points to sporadic transcriptional bursting as an alternative explanation for the previ- ous  FISH  results. Why the  burst  interval should   be  longer  for  Nanog   than  other pluripotency factors expressed at similar mRNA levels is unknown.

Leaving  aside  disputes over  construct design,  the  real  issue  is  whether  Nanog heterogeneity in ESCs under certain condi- tions  has   biological   meaning.    Without feeders  or  2i,  ESCs   in  serum  and  LIF show  variegated  expression not  only of Nanog  but  also  of several other  pluripo- tency transcription  factors.  These factors, such as Klf4, Esrrb, and Rex1, are typically downregulated  at the onset of ESC differ- entiation,    during    implantation    in   the embryo, and in cultured  postimplantation epiblast stem  cells (EpiSCs) (Nichols and Smith, 2012). This observation, along with the readily detected upregulation of early differentiation    markers,    suggests   that feeder-free   ESCs    in   serum    and    LIF comprise  both  self-renewing  stem   cells and a spectrum of cells in transition toward differentiation  (Marks  et  al.,  2012).  The conflicting stimuli provided  by serum may promote disorder, while the potent activity of LIF as both a self-renewal and a reprog- ramming  signal  (Yang  et  al.,  2010)  may induce reversion during transition.  How- ever, it should be noted that a substantial proportion of Nanog low cells are destined for differentiation  and  shedding from the culture    during    passaging    (Chambers et al., 2007). Thus, the ESC heterogeneity that  has been  documented may  well  be primarily  a consequence of a  disordered signaling  environment  created  by  a  spe- cific set of in vitro conditions. This culture specificity  raises  questions about  overall functional significance. Nanog expression in the very early embryo appears to fluc- tuate  stochastically. However,  that  form of heterogeneity  precedes  emergence of the  pluripotent  epiblast,  in which  Nanog expression is consolidated and from which ESCs  are  actually  derived  (Nichols  and Smith, 2012). Importantly, there is currently no evidence that either fluctuating expres-  sion of pluripotency  factors or state  rever- sion  occurs   during  epiblast   progression and lineage commitment in vivo. Instead, the  variation  that  many  have  observed may simply be a culture epiphenomenon: an attractive playbox for experimentalists and   modellers, but with questionable relevance  for the way in which pluripotent cells really make fate decisions.