PERSPECTIVES

are caused by loss of some of the synMuv of transcription and chromatin-regulating References B chromatin regulators (5 , 7 , 8 ). Notably, factors. A detailed understanding of these 1. B. Tursun, T. Patel, P. Kratsios, O. Hobert, Science 311, 304 (2010); 10.1126/science.1199082. although LIN-53 is a member of the synMuv would allow for precise and directed manip- 2. O. Uchida et al., Development 130, 1215 (2003). B group, some experiments (8 ) suggest that it ulation, potentially enabling the creation of 3. S. Sarin, C. Antonio, B. Tursun, Development 136, 2933 is not among the synMuv B proteins whose any cell type from any other. Although there (2009). loss causes a strong soma-to-germ transfor- are some examples of reprogramming cul- 4. D. S. Fay, J. Yochem, Dev. Biol. 306, 1 (2007). 5. S. Strome, R. Lehmann, Science 316, 392 (2007). mation. Figuring out which chromatin regu- tured mammalian cells from one cell type 6. R. Ciosk et al., Science 311, 851 (2006). lators limit competence to respond to lineage- to another, investigators are far from having 7. Y. Unhavaithaya et al., Cell 111, 991 (2002). regulating factors, and in which cell types complete control over the process (9 –12 ). 8. D. Wang et al., Nature 436, 593 (2005). 9. K. Takahashi, S. Yamanaka, Cell 126, 663 (2006). (see the fi gure), is an important future direc- Studies such as this can reveal fundamental 10. Q. Zhou, D. A. Melton, Cell Stem Cell 3, 382 (2008). tion for research. aspects of the processes that keep cell types 11. T. Graf, T. Enver, Nature 462, 587 (2009). Tursun et al. give a big boost to cellular distinct, and reveal important targets for 12. B. Feng, J. H. Ng, J. C. Heng, H. H. Ng, Cell Stem Cell 4, reprogramming efforts. Each cell type of regenerative medicine approaches that seek 301 (2009). the body might be governed by a unique set to step over barriers between cell types. 10.1126/science.1201288

ASTRONOMY

The linear isomer of C3H2 has been proposed Disclosing Identities in Diffuse as the source of broad absorption bands in diffuse interstellar clouds, but it is still Interstellar Bands being debated. Takeshi Oka1 and Benjamin J. McCall2 on January 28, 2011 ust as detectives use fi ngerprints to pin The strongest DIB, at the wavelength the and us. More than that is not known down criminals, astronomers use spec- of 4430 Å, was initially observed by Annie with certainty. It is diffi cult to hunt criminals Jtra to identify molecules in interstellar Jump Cannon, the fi rst astronomer to clas- by using blurry fi ngerprints. In comparison, space. Astronomical spectroscopy, especially sify stars systematically by spectroscopy, radio detections of molecules are clear-cut spectroscopy of molecular ions, is develop- sometime between 1911 and 1919 (3 –5 ). because their fi ngerprints (rotational spectra) ing rapidly. Numerous interstellar negative The carriers of DIBs are molecules, not are extremely sharp. Milestone discoveries + + ions, the very high abundance of H3 in the atoms or solid grains, and they are not in the of molecules, like H2O, H2CO, CO, HCO , Galactic center, and the extremely strong stars but are in huge diffuse clouds between and HC5N, were initially claimed by detect- far-infrared absorptions of OH+, ing one spectral line, and there www.sciencemag.org + + + A H2O , H2Cl , and CH have all have not been any mistakes. been discovered during the last Numerous hypotheses, many few (1 ). In contrast to these from eminent spectroscopists and rapid developments in the dis- astronomers, have been proposed covery of new molecular spec- to explain DIBs, and most of tra, whose sharp features can be these hypotheses look far-fetched

thought of as clean fi ngerprints, in hindsight. The chemist Bill Downloaded from there exists a group of several Klemperer once said that “there hundred intriguing broad opti- is no better way to lose a scientifi c cal spectra (see the fi gure, panel B reputation than to speculate on A) called the diffuse interstellar the carrier for the diffuse bands” bands (DIBs). These blurry fi n- (6 ). During the past 10 years, can- − − gerprints have defied attempts didates have included C7 , C3H2 , + + by many astronomers, physi- C10H8 , and HC4H , proposed cists, and chemists to understand on the bases of their laboratory them for many decades. Maier et spectra. However, they have not al. (2 ) now have a suspect in cus- been accepted because their fi n- tody as the molecule responsible gerprints were not a good enough Y for DIBs—the linear carbene match to DIBs. There have been molecule, l-C H —but more many more proposals not based

Y SURVE 3 2 evidence will be needed to get a on laboratory spectra, but these L SK conviction. are more speculative, and akin TICA The basis of blurry fi ngerprints. (A) Diffuse interstellar bands (DIBs) are broad visible spectral features caused by molecules in interstellar matter absorbing - to naming a suspect based on cir- AR OP 1 Department of Chemistry, University of light. (B) The star HD 183143 at the center of this starfi eld in the cumstantial evidence. Chicago, Chicago, IL 60637, USA. 2Depart- ALOM is used to measure DIBs. The solid curves show DIBs for the wavelength Maier et al.’s proposal that : P ment of Chemistry, University of Illinois, T Urbana, IL 61801, USA. E-mail: t-oka@ ranges from 4812 to 4956 Å, 5431 to 5472 Å, and 6144 to 6176 Å. The simulations two DIBs at 4881 Å and 5450 Å EDI R

C of Maier et al. based on their laboratory data for l-C H are shown as dashed lines. l uchicago.edu; [email protected] 3 2 are caused by -C3H2 has stirred

www.sciencemag.org SCIENCE VOL 331 21 JANUARY 2011 293 Published by AAAS PERSPECTIVES

l great interest within the astronomy commu- high abundance of -C3H2 toward the star the broad bands have a constant intensity nity. Maier is a leader in the fi eld of laboratory HD 183143 (center of panel B of the fi gure). ratio in different clouds. It may also be pos- l spectroscopic astrophysics and, using a vari- This line of sight, which has been famous sible to search for -C3H2 in diffuse clouds ety of newly developed techniques, has been in the DIB community since the pioneering through its pure-rotational transitions (16 ); amazingly productive in discovering new vis- work of Herbig (11 ), is remarkable in that it however, a direct comparison of absorption ible and ultraviolet spectra of carbon-contain- contains very few carbon-chain molecules, in the optical and the radio spectrum would 12 13 ing unstable molecules, which are likely car- such as C2 ( ) or C3 ( ). In this respect, require a background source that is bright at riers of DIBs (only the almost collision-free HD 183143 is an extreme case; its polar both wavelengths. environment of interstellar space allows such opposite is the star HD 204827, which has References and Notes molecules to exist in abundance there; in the by far the highest C3 and C2 column densities lab, they are diffi cult to prepare). This par- of all sightlines so far studied and led to the 1. Those developments were hotly debated at a recent sym- posium, “Spectroscopy of molecular ions in the labora- ticular candidate molecule had already been discovery by our co-workers and ourselves tory and in (extraterrestrial) space,” held in Kos, Greece,

observed by radio emission in dense inter- of a group of DIBs called “C2 DIBs” whose 3 to 6 October 2010; http://fermi.uchicago.edu/~smiles/. 2. J. P. Maier et al., Astrophys. J. 726, 41 (2011). stellar clouds (~100 times denser than diffuse intensities correlate well with the C2 column 12 3. G. H. Herbig, Astrophys. J. 196, 129 (1975). clouds where DIBs are observed) nearly 20 densities ( ). These two sightlines have 70 7 4. A. D. Code, Publ. Astron. Soc. Pac. , 407 (1958). years ago ( ). Cyclic C3H2 has been observed also served as contrasting prototypes for 5. G. E. Bromage, Q. J. R. Astron. Soc 28, 295 (1987). 8 l 14 15 6. E. Herbst, Annu. Rev. Phys. Chem. 46, 27 (1995). in diffuse clouds ( ), so the presence of -C3H2 detailed compilations of the DIBs ( , ). may not be surprising. Related molecules If Maier’s claim about l-C H is correct, this 7. J. Cernicharo et al., Astrophys. J. 368, L39 (1991). 3 2 358 such as C , C H, and C have also been seen molecule must be more than two orders of 8. R. Lucas, H. S. Liszt, Astron. Astrophys. , 1069 (2000). 2 2 3 9. P. Birza, A. Chirakolava, M. Araki, P. Kolek, J. P. Maier, J. in diffuse clouds, and such a simple molecule magnitude more abundant than C2 and C3 in Mol. Spectrosc. 229, 276 (2005). is more appealing than very complicated mol- HD 183143. Although the chemical mecha- 10. E. Achkasova, M. Araki, A. Denisov, J. P. Maier, J. Mol. ecules whose spectra are diffi cult to obtain in nism for producing hydrocarbon molecules Spectrosc. 237, 70 (2006). 11. G. H. Herbig, Z. Astrophys. 64, 512 (1966). the laboratory. in diffuse clouds is not well understood, it is 12. J. A. Thorburn et al., Astrophys. J. 584, 339 (2003). Maier’s group had previously recorded hard to imagine such a tremendous discrep- 13. T. Oka et al., Astrophys. J. 582, 823 (2003). 680

l l 14. L. M. Hobbs et al., Astrophys. J. , 1256 (2008). on January 28, 2011 sharp fi ngerprints of -C3H2 in the laboratory ancy in the abundance of C3 and -C3H2. in the 6150 to 6330 Å region, which corre- At present, we conclude that there is 15. L. M. Hobbs et al., Astrophys. J. 705, 32 (2009). l 16. J. Cernicharo, P. Cox, D. Fossé, R. Güsten, Astron. Astro- spond to this molecule’s lowest-energy elec- insuffi cient evidence to “convict” -C3H2 as phys. 351, 341 (1999). tronic transition (9 , 10 ). The major break- a carrier of the DIBs. Searches for both the 17. We thank L. M. Hobbs, P. Sonnentrucker, D. E. Welty, and l D. G. York for helpful discussions. through in the present work was the proof that sharp and broad -C3H2 bands in a number two much broader bands at 4881 and 5450 Å of diffuse clouds could establish whether the are also caused by a higher electronic transi- sharp bands are a perfect match, and whether 10.1126/science.1200144 l tion of -C3H2. The broad spectral lines of a l simple molecule like -C3H2 appear to be the result of the extremely short lifetime (~3 × MICROBIOLOGY www.sciencemag.org 10−13 s) of the second electronic excited state. This interpretation follows from the uncer- tainty principle—a shorter transition time Another Microbial Pathway for corresponds to higher uncertainty in energy and hence a broader spectral line. The match of the two broad bands to DIBs Acetate Assimilation

seems reasonable, but with such “blurry fi n- Scott A. Ensign Downloaded from gerprints,” it does not constitute proof beyond a reasonable doubt. In panel B of the fi gure, In a salt-loving Archaeon, a patchwork of enzymes creates a variation of the glyoxylate cycle. the interstellar spectra (solid curves) and sim- ulations from laboratory data (dashed curves) ountless students in introductory the glyoxylate cycle, and it allows acetyl- are shown, with the 4881 Å DIB at left, and biochemistry classes have heard CoA to be used as a replenishing source for the 5450 Å DIB at center. Maier et al. also Cthe adage “fats burn in the fl ame of carbon in the synthesis of glucose and other l 1 searched for the sharp transitions of -C3H2 carbohydrates.” It refers to the inability of important reactions ( ). However, a num- in the 6150 to 6330 Å region; the match with vertebrates to convert acetyl coenzyme A ber of acetate-using microorganisms lack their spectra of interstellar clouds is promis- (acetyl-CoA), an important metabolic mol- one of the signature enzymes involved in ing, but not entirely convincing (panel B of the ecule, into intermediate compounds that the glyoxylate cycle, isocitrate lyase, dem- fi gure, lower right). Unfortunately, different replenish carbon in the citric acid cycle and onstrating that other pathway(s) for acetate stars were used in the searches for the broad enable the synthesis of glucose. In contrast, assimilation must exist (2 – 4). In 2007, fully and sharp bands, so it remains to be seen plants and certain bacteria, fungi, and inver- 50 years after the discovery of the glyox- l whether all of the supposed -C3H2 transitions tebrates have solved this barrier to “anabo- ylate cycle, investigators revealed the fi rst appear in a single diffuse cloud, and whether lism from acetate” by using a variation of complete details of one of these alternate the intensities of all these transitions have a the citric acid cycle. This variation is called pathways (5 ). Now, on page 334 of this constant ratio in all clouds (as they should if issue, Khomyakova et al. (6 ) describe yet they are caused by the same molecule). another acetate-assimilation pathway. It is et Department of Chemistry and Biochemistry, Utah State Uni- Another aspect of the work by Maier versity, Logan, UT 84322–0300, USA. E-mail: ensigns@ in a salt-loving (halophilic) microorganism al. that is diffi cult to accept is the apparently cc.usu.edu belonging to the Archaea, and incorporates

294 21 JANUARY 2011 VOL 331 SCIENCE www.sciencemag.org Published by AAAS