26.4 Catalytic RNAs and the RNA World Hypothesis in Chapter 26 RNA Metabolism

26.4 Catalytic RNAs and the RNA World Hypothesis

The study of posttranscriptional processing of RNA molecules led to one of the most exciting discoveries in biochemistry — the existence of RNA enzymes or ribozymes. The best-characterized ribozymes are the self-splicing group I introns, RNase P, and the hammerhead ribozyme (discussed below). Most of the activities of these ribozymes are based on two fundamental reactions: transesterification (Fig. 26-14) and phosphodiester bond hydrolysis (cleavage). The substrate for a ribozyme is often an RNA molecule, and it may even be part of the ribozyme itself. When its substrate is RNA, the RNA catalyst can make use of base-pairing interactions to align the substrate for the reaction.

Ribozymes Share Features with Protein Enzymes

Like protein enzymes, ribozymes vary greatly in size. A self-splicing group I intron may have more than 400 nucleotides. In comparison, the hammerhead ribozyme consists of two RNA strands with a total of just 41 nucleotides. Also as with protein enzymes, the three-dimensional structure of ribozymes is important for function. Ribozymes, like protein enzymes, are inactivated by heating above their melting temperature or by the addition of denaturing agents or complementary oligonucleotides, which disrupt normal base-pairing patterns. Ribozymes can also be inactivated if essential nucleotides are changed. The secondary structure of a self-splicing group I intron from the 26S rRNA precursor of Tetrahymena is shown in detail in Figure 26-37. This secondary structure highlights the large number of base-pairing and other noncovalent interactions that must occur for the ribozyme to adopt a catalytic structure. Just as amino acid mutations can change activities of protein enzymes, nucleotide mutations can alter the noncovalent interactions required for ribozyme folding and catalysis.

A two-part figure shows the secondary structure of a self-splicing r R N A intron and part b shows the three-dimensional structure of a reaction intermediate of the same intron. — FIGURE 26-37 Secondary structure of the self-splicing rRNA intron of *Tetrahymena*. (a) A two-dimensional representation of secondary structure immediately prior to the initiation of the reaction. Intron sequences are shaded yellow and light red; flanking exon sequences are green; the internal guide sequences that help to align reacting segments at the active site are purple. Each thin, light-red line represents a bond between adjacent nucleotides in a continuous sequence (a device necessitated by showing this complex molecule in two dimensions). Short blue lines represent normal base pairing; blue dots indicate G–U base pairs. All nucleotides are shown. The catalytic core of the self-splicing activity is shaded in gray. Some base-paired regions are labeled (P1, P3, P2.1, P5a, and so forth) according to an established convention for this RNA molecule. The P1 region, which contains the internal guide sequence (purple), is the location of the $5^{'}$ $5 prime$ splice site (black arrow). Part of the internal guide sequence pairs with the end of the $3^{'}$ $3 prime$ exon, bringing the $5^{'}$ $5 prime$ and $3^{'}$ $3 prime$ splice sites (black arrows) into close proximity. (b) Three-dimensional structure of a reaction intermediate of the same intron, after guanosine-mediated cleavage (Fig. 26-14) and prior to exon ligation. Segments are colored as in (a). [Data from (a) PDB ID 1GID, J. H. Cate et al., *Science* 273:1678, 1996. (b) PDB ID 1U6B, P. L. Adams et al., *Nature* 430:45, 2004.]

FIGURE 26-37 Secondary structure of the self-splicing rRNA intron of *Tetrahymena*. (a) A two-dimensional representation of secondary structure immediately prior to the initiation of the reaction. Intron sequences are shaded yellow and light red; flanking exon sequences are green; the internal guide sequences that help to align reacting segments at the active site are purple. Each thin, light-red line represents a bond between adjacent nucleotides in a continuous sequence (a device necessitated by showing this complex molecule in two dimensions). Short blue lines represent normal base pairing; blue dots indicate G–U base pairs. All nucleotides are shown. The catalytic core of the self-splicing activity is shaded in gray. Some base-paired regions are labeled (P1, P3, P2.1, P5a, and so forth) according to an established convention for this RNA molecule. The P1 region, which contains the internal guide sequence (purple), is the location of the $5^{'}$ $5 prime$ splice site (black arrow). Part of the internal guide sequence pairs with the end of the $3^{'}$ $3 prime$ exon, bringing the $5^{'}$ $5 prime$ and $3^{'}$ $3 prime$ splice sites (black arrows) into close proximity. (b) Three-dimensional structure of a reaction intermediate of the same intron, after guanosine-mediated cleavage (Fig. 26-14) and prior to exon ligation. Segments are colored as in (a). [Data from (a) PDB ID 1GID, J. H. Cate et al., *Science* 273:1678, 1996. (b) PDB ID 1U6B, P. L. Adams et al., *Nature* 430:45, 2004.]

On the left in part a, a yellow vertical piece has a loop at the bottom labeled p 5 b that is mostly connected by bonds. Above, the left-hand strand bends left to form a loop labeled P 5 c before bending back for another vertical section bonded to the right-hand piece, which is connected to the lower piece by a rod. A rod connects this right-hand piece to another vertical piece above labeled P 5 a. On the left, there are two loops adjacent to the rod. Rods connect the top of the left-hand piece with the top of a vertical piece to the right labeled P 5 and the top of the right-hand piece with the top of the right-hand piece of P 5. This vertical piece is in a gray rectangle. It has a region where the sides separate to form an opening, then another vertical piece labeled P 4. A rod connects the left-hand piece to a vertical piece below labeled P 6, which emerges from the gray region to form a loop and then reach a vertical piece labeled P 6 a above a loop above a vertical piece labeled P 6 b above a final loop at the bottom. Between P 4 an P 6, orange bars crosses over to join the lower right-hand piece of P 4 to a bottom horizontal line and the upper right-hand piece of P 6 to an upper horizontal line. These lines run to the right and the top bar joins the upper right end of a right side vertical piece labeled P 9.0 while the bottom piece joins the upper left end of a middle right side vertical piece labeled P 3. Above these horizontal lines, a tan vertical piece labeled P 1 has a loop at the top that extends beyond the gray area with a blue bond to a green piece to the right that ends at 39. The lower left of P 1 is purple and labeled internal guide sequence. It has a bar that runs left and then down to the upper left of a tan vertical sequence labeled P 2 a the bottom center that ends with a loop at its bottom. The lower right of P 1 is green and ends at 59. An arrow points to the place where the green area meets the beige are above. From the upper right of P 2, a bar connects to the lower left of a vertical piece labeled P 2.1 that ends in a loop above. Its lower right end has a bar that runs right then up to join to the lower left of the left-hand side of P 3. P 3 has a bar from its lower right joining to tan P 8 below, which ends in a loop and has an upper right sie connected to a bar connected to a thicker region that runs into the gray rectangle to join to the lower left of P 9.0. P 3 has a loop near its top just where it enters the gray rectangle. P 9.0 above has a left side connected by a bar to a tan piece connected by two bonds to a tan piece to the left that forms a “U” shape. An arrow points to the lower right corner, where it changes from tan to green, and it ends at 39. The top of this tan piece crosses a bar from the tan piece to the right to join a tan bar that runs up to a horizontal piece labeled P 9.2 with a loop in its center. P 9.2 joins to a vertical piece with P 9.1 at the bottom, a loop, and then P 9.1 a above beneath a terminal loop. P 9.1 also joins to a tan horizontal piece that extends to the left labeled P 9. Part b shows a structure with a yellow double helix labeled P 6 at the lower right that runs up at a slight diagonal and ends at P 5 extending up above the main structure. A tannish pink helix begins at P 8 to the lower left and extends roughly diagonally to the upper right with two protrusions to the left with P 2 below P 1. A green piece runs from the right end of P a to the yellow part in the center. A purple region labeled internal guide sequence extends left above P 1 and is labeled P 10. There is a green loop above it. The upper right of the tannish pink part is labeled P 9, with P 9.0 below and P 7 at the center right. P 3 is shown where the pink and yellow parts meet.

Ribozymes share several properties with enzymes besides accelerating the reaction rate, including kinetic behavior and specificity. Binding of the guanosine cofactor to the Tetrahymena group I rRNA intron is saturable ( $K_{m} < 30 μ M$ $upper K Subscript m Baseline less-than 30 mu upper M$ ) and can be competitively inhibited by $3^{'}$ $3 prime$ -deoxyguanosine. The intron is very precise in its excision reaction, largely due to a segment called the internal guide sequence that can base-pair with exon sequences near the $5^{'}$ $5 prime$ splice site (Fig. 26-37). This pairing promotes the alignment of specific bonds to be cleaved and rejoined.

Because the intron itself is chemically altered during the splicing reaction — its ends are cleaved — it may seem to lack one key enzymatic property: the ability to catalyze multiple reactions. Closer inspection has shown that after excision, the 414 nucleotide intron from Tetrahymena rRNA can, in vitro, act as a true enzyme (but in vivo it is quickly degraded). A series of intramolecular cyclization and cleavage reactions in the excised intron leads to the loss of 19 nucleotides from its $5^{'}$ $5 prime$ end. The remaining 395 nucleotide, linear RNA — referred to as L-19 IVS (intervening sequence) — promotes nucleotidyl transfer reactions in which some oligonucleotides are lengthened at the expense of others (Fig. 26-38). The best substrates are oligonucleotides, such as a synthetic ${(C)}_{5}$ $left-parenthesis upper C right-parenthesis Subscript 5$ oligomer, that can base-pair with the same guanylate-rich internal guide sequence that held the $5^{'}$ $5 prime$ exon in place for self-splicing.

A two-part figure shows the in vitro catalytic activity of L-19 I V S by showing the generation of the structure in part a and the steps involved in catalytic activity in part b. — FIGURE 26-38 In vitro catalytic activity of L-19 IVS. (a) L-19 IVS (intervening sequence, the term once used for intron) is generated by the autocatalytic removal of 19 nucleotides from the $5^{'}$ $5 prime$ end of the spliced *Tetrahymena* intron. The cleavage site is indicated by the arrow in the internal guide sequence (boxed). The G residue (shaded) added in the first step of the splicing reaction is part of the removed sequence. A portion of the internal guide sequence remains at the $5^{'}$ $5 prime$ end of L-19 IVS. (b) L-19 IVS lengthens some RNA oligonucleotides at the expense of others in a cycle of transesterification reactions (steps through ). The $3^{'}$ $3 prime$ OH of the G residue at the $3^{'}$ $3 prime$ end of L-19 IVS plays a key role in this cycle (note that this is *not* the G residue added in the splicing reaction). ${(C)}_{5}$ $left-parenthesis upper C right-parenthesis Subscript 5$ is one of the ribozyme’s better substrates because it can base-pair with the guide sequence remaining in the intron.

Part a shows a spliced r R N A intron. From left to right, it is (5 prime) red highlighted G A A A U A G C A A U A U U then a box containing U A C C U, then an arrow pointing down to the right of this U within the box, then another U followed by U G G A G G G then A outside of the box followed by a yellow piece intersected by wavy vertical lines followed by a yellow piece ending at G bonded to O H (3 prime). An arrow points downward accompanied by a branched arrow showing the loss of 19 nucleotides from the 5 prime end. This yields L-19 I V S, which is (5 prime) then a box with U U G G A G G G then A outside of the box followed by a yellow piece intersected by wavy vertical lines followed by a yellow piece ending at G bonded to O H (3 prime). Part b shows a series of four steps showing the catalytic activity of L-19 I V S. (5 prime) U U G G A G G G A is connected to a yellow strand that bends back to the upper left. It is intersected by two wavy lines before continuing on to end at G bonded to OH (3 prime). An arrow labeled 1 points right and a curved line shows the addition of (C) subscript 5 end subscript above red highlighted H O bonded to C C C C C. This yields a similar structure in which the red highlighted nucleotides have formed bonds. U U is unpaired, then G is bonded to red highlighted C, G is bonded to red highlighted C, A is unbonded, and G is bonded to red highlighted C. At the 3 prime end of the molecule, a red pair of electrons is shown on O H with an arrow pointing to the space between the first C and the second C, which is hydrogen bonded to C. An arrow pointing downward is labeled 2 and accompanied by a curved arrow showing the loss of red highlighted H O bonded to C C C C, nonhighlighted (C) subscript 4 end subscript. This yields U U G G A G G G A is connected to a yellow strand that bends back to the upper left. It is intersected by two wavy lines before continuing on to end at G bonded to red highlighted C connected by a red bond to red highlighted OH. An arrow pointing left is labeled 3 and is accompanied by a curved line showing the addition of blue highlighted H O bonded to C C C C C, nonhighlighted (C) subscript 5 end subscript. This yields a similar molecule in which the blue highlighted bases have formed hydrogen bonds. The sequence is U U G bonded to blue highlighted C, G bonded to blue highlighted C, A not bonded to blue highlighted C above, G bonded to blue highlighted C, then G not bonded to blue highlighted C. A red pair of electrons on blue highlighted O H has a red arrow pointing to C bonded to G and to red highlighted O H at the 3 prime end. An arrow pointing upward is labeled 4 and accompanied by a branched arrow showing the loss of blue highlighted C C C C C red highlighted C connected by a red bond to red highlighted O H, nonhighlighted (C) subscript 6 end subscript. This yields the starting material ready to repeat the process.

The enzymatic activity of the L-19 IVS ribozyme results from a cycle of transesterification reactions mechanistically similar to self-splicing. Each ribozyme molecule can process about 100 substrate molecules per hour and is not altered in the reaction — that is, the intron acts as a catalyst. It follows Michaelis-Menten kinetics, is specific for RNA oligonucleotide substrates, and can be competitively inhibited. The $k_{cat} / K_{m}$ $k Subscript cat Baseline slash upper K Subscript m Baseline$ (specificity constant) is $10^{3} M^{- 1} s^{- 1}$ $10 cubed upper M Superscript negative 1 Baseline s Superscript negative 1$ , lower than that of many protein enzymes, but the ribozyme accelerates hydrolysis by a factor of $10^{10}$ $10 Superscript 10$ relative to the uncatalyzed reaction. It makes use of substrate orientation, covalent catalysis, and metal-ion catalysis — all strategies shared with protein enzymes.

Ribozymes Participate in a Variety of Biological Processes

E. coli RNase P has both an RNA component (the M1 RNA, with 377 nucleotides) and a protein component ( $M_{r} 17, 500$ $upper M Subscript r Baseline 17 comma 500$ ). In 1983, Sidney Altman and Norman Pace and their coworkers discovered that under some conditions, the M1 RNA alone is capable of catalysis, cleaving tRNA precursors at the correct position. The protein component apparently serves to stabilize the RNA or facilitate its function in vivo. The RNase P ribozyme recognizes the three-dimensional shape of its pre-tRNA substrate, along with the CCA sequence, and thus can cleave the $5^{'}$ $5 prime$ leaders from diverse tRNAs (Fig. 26-25).

The known catalytic repertoire of ribozymes continues to expand. Some virusoids, small RNAs associated with plant RNA viruses, include a structure that promotes a self-cleavage reaction; the small hammerhead ribozyme illustrated in Figure 26-39 is in this class, catalyzing the hydrolysis of an internal phosphodiester bond. There are at least nine structural classes of ribozymes that engage in self-cleavage; all use general acid and base catalysis (Fig. 6-8) to promote the attack of a $2^{'}$ $2 prime$ -hydroxyl group on an adjacent phosphodiester bond. Despite being surrounded by proteins, the splicing reaction that occurs in a spliceosome relies on a catalytic center formed by the U2, U5, and U6 snRNAs and intron (see Figs 26-16 and 26-17). And, as we shall see in Chapter 27, an RNA component of ribosomes catalyzes the synthesis of proteins. Exploring catalytic RNAs has provided new insights into catalytic function in general and has important implications for our understanding of the origin and evolution of life on this planet.

A two-part figure shows a hammerhead ribozyme with part a showing the sequences and part b showing the three-dimensional structure. — FIGURE 26-39 Hammerhead ribozyme. Some virusoid RNAs include small segments that promote site-specific RNA cleavage reactions associated with replication. These segments are called hammerhead ribozymes, because their secondary structures are shaped like the head of a hammer. (a) The minimal sequences required for catalysis by the ribozyme. The boxed nucleotides are highly conserved and are required for catalytic function. Guanine nucleotides shaded pink form part of the active site. The arrow indicates the site of self-cleavage. (b) Three-dimensional structure of a hammerhead ribozyme (see Fig. 8-25b for a view of another hammerhead ribozyme). The strands are colored as in (a). The hammerhead ribozyme is a metalloenzyme; ${Mg}^{2 +}$ $Mg Superscript 2 plus$ ions are required for activity in vivo. The phosphodiester bond at the site of self-cleavage is indicated by an arrow. [Data from PDB ID 3ZD5, M. Martick and W. G. Scott, *Cell* 126:309, 2006.]

FIGURE 26-39 Hammerhead ribozyme. Some virusoid RNAs include small segments that promote site-specific RNA cleavage reactions associated with replication. These segments are called hammerhead ribozymes, because their secondary structures are shaped like the head of a hammer. (a) The minimal sequences required for catalysis by the ribozyme. The boxed nucleotides are highly conserved and are required for catalytic function. Guanine nucleotides shaded pink form part of the active site. The arrow indicates the site of self-cleavage. (b) Three-dimensional structure of a hammerhead ribozyme (see Fig. 8-25b for a view of another hammerhead ribozyme). The strands are colored as in (a). The hammerhead ribozyme is a metalloenzyme; ${Mg}^{2 +}$ $Mg Superscript 2 plus$ ions are required for activity in vivo. The phosphodiester bond at the site of self-cleavage is indicated by an arrow. [Data from PDB ID 3ZD5, M. Martick and W. G. Scott, *Cell* 126:309, 2006.]

Part a shows a vertical structure with the 5 prime end at the lower left and the 3 prime end at the lower right. It begins at red highlighted G at the 5 prime end bonded to red highlighted C at the 3 prime end. The 5 prime strand runs up as light green G A U, then widens to form an opening with G U A C unpaired before coming back to form a vertical paired region of U A C C A G. It then bends to the left with a series of bases shown in a box. These are C U G A, with A at the left side. U is at the upper left, then another box starts with pink G then a then G that begins a vertical sequence of paired bases that continues outside of the box with U C C. Next, C A A A U A form a loop, then G G A pair with the sequence to the left. C in a box is paired with G in the box to the left and then the strand bends right to pink G A A and finally A paired to U in a dark green strand below. This begins a horizontal sequence of C G C blue highlighted C 3 prime that is paired with a strand below of 5 prime blue G bonded to the blue C, then G C G U. The strand bends down to a yellow C, then to a purple C bonded to G at the top of a vertical sequence of bonded nucleotides. The sequence continues as U G G U A before looping out with U C C A and then returning to end with a vertical sequence of A U C light red C at the 3 prime end. The overall structure is mostly linear at the bottom with a single loop and with a wider loop at the top with a strand to the right and a stem loop on top labeled hammerhead. Part b shows a structure with a red 5 prime end next to a red 3 prime end at the lower left. The 5 prime end begins a strand that coils right and then left while the 3 prime end begins dark green strand that curves left and then diagonally up past a purple C and an arrow, then past a yellow C and pink G to end at a blue 5 prime end. The blue 3 prime and 5 prime ends are at the upper left. Where the dark green strand crosses the light green strand just before the 5 prime end, there is a pink region above pink G next to yellow C next to pink G next to purple C. The overall structure is rounded.

Ribozymes Provide Clues to the Origin of Life in an RNA World

The extraordinary complexity and order that distinguish living from inanimate systems are key manifestations of fundamental life processes. Maintaining the living state requires that selected chemical transformations occur very rapidly — especially those that use environmental energy sources and synthesize elaborate or specialized cellular macromolecules. Life depends on powerful and selective catalysts — enzymes — and on informational systems capable of both securely storing the blueprint for these enzymes and accurately reproducing the blueprint for generation after generation. Chromosomes encode the blueprint not for the cell but for the enzymes that construct and maintain the cell. The parallel demands for information and catalysis present a classic conundrum: what came first, the information needed to specify structure or the enzymes needed to maintain and transmit the information?

A photo of Carl Woese in a laboratory. — Carl Woese, 1928–2012

How might a self-replicating polymer come to be? How might it maintain itself in an environment where the precursors for polymer synthesis are scarce? How could evolution progress from such a polymer to the modern DNA-protein world? These difficult questions can be addressed by careful experimentation, providing clues about how life on Earth began and evolved.

The unveiling of the structural and functional complexity of RNA led Carl Woese, Francis Crick, and Leslie Orgel to propose in the 1960s that this macromolecule might serve as both information carrier and catalyst. Since that time, at least six lines of evidence have given increasing substance to their RNA world hypothesis.

1. Prebiotic Chemistry Experiments

The probable origin of purine and pyrimidine bases is suggested by experiments designed to test hypotheses about prebiotic chemistry (pp. 31–32). Beginning with simple molecules thought to be present in the early atmosphere ( ${CH}_{4}$ $CH Subscript 4$ , ${NH}_{3}$ $NH Subscript 3$ , $H_{2} O$ $upper H Subscript 2 Baseline upper O$ , $H_{2}$ $upper H Subscript 2$ ), electrical discharges mimicking lightning generate, first, more reactive molecules such as HCN and aldehydes, then an array of amino acids and organic acids (see Fig. 1-33). When molecules such as HCN become abundant, purine and pyrimidine bases are synthesized in detectable amounts. Remarkably, a concentrated solution of ammonium cyanide, refluxed for a few days, generates adenine in yields of up to 0.5% (Fig. 26-40). Adenine may well have been the first and most abundant nucleotide constituent to appear on Earth. Intriguingly, most enzyme cofactors contain adenosine as part of their structure, although it plays no direct role in the cofactor function (see Fig. 8-41). This may suggest an evolutionary relationship. Based on the simple synthesis of adenine from cyanide, adenine may simply have been abundant and available.

A figure shows the synthesis of adenine from ammonium cyanide. — FIGURE 26-40 Experiments supporting prebiotic synthesis of adenine from ammonium cyanide. Adenine is derived from five molecules of cyanide, denoted by shading.

H C N (N H 4 C N) is shown with an arrow pointing right above the word reflux. This yields a five-membered ring that shares its right side with a six-membered ring. The left-hand five-membered ring has N substituted for C at position 9 at the lower left vertex and bonded to H enclosed within a light red box along with the lower left bond to C at the left side vertex at position 8. There is a double bond between positions 7 and 8. N is substituted for C at position 7 and enclosed in a light red box that contains the upper right bond from positions 8 to 5 and C at position 5. There is a double bond between positions 5 and 6 that is shared by the two rings. C at position 4 is enclosed a light red box that includes the lower left bond to the six-membered ring and N substituted for C at position 3 at the bottom vertex. There is a double bond at the lower right side between positions 2 and 3. N is substituted for C at position 1 and enclosed in a light red box along with the bond to C at position 2 below. There is a double bond between positions 5 and 6. C at position 6 is enclosed in a light red box with a bond above to N H 2 in the same box.

2. The Existence of Catalytic RNAs

In an “RNA world,” RNAs, not proteins, act as catalysts. Perhaps more than anything else, the discovery of ribozymes gave life to the RNA world hypothesis and led to widespread speculation that an RNA world might have been important in the transition from prebiotic chemistry to life (see Fig. 1-35). The parent of all life on this planet, in the sense that it could reproduce itself across the generations from the origin of life to the present, might have been a self-replicating RNA, or a polymer with equivalent chemical characteristics.

3. The Expanding Catalytic Repertoire of Ribozymes

A self-replicating polymer would quickly use up available supplies of precursors provided by the relatively slow processes of prebiotic chemistry. Thus, from an early stage in evolution, metabolic pathways would be required to generate precursors efficiently, with the synthesis of precursors presumably catalyzed by ribozymes. The extant ribozymes found in nature have a limited repertoire of catalytic functions, and of the ribozymes that may once have existed, no trace is left. To explore the RNA world hypothesis more deeply, we need to know whether RNA has the potential to catalyze the many different reactions needed in a primitive system of metabolic pathways.

The search for RNAs with new catalytic functions has been aided by the development of a method that rapidly searches pools of random polymers of RNA and extracts those with particular activities; known as SELEX, this is nothing less than accelerated evolution in a test tube (Box 26-4). It has been used to generate RNA molecules that bind to amino acids, organic dyes, nucleotides, cyanocobalamin, and other molecules. Researchers have isolated ribozymes that catalyze ester and amide bond formation, $S_{N} 2$ $upper S Subscript upper N Baseline 2$ reactions, metallation of (addition of metal ions to) porphyrins, and carbon–carbon bond formation. The evolution of enzymatic cofactors with nucleotide “handles” that facilitate their binding to ribozymes might have further expanded the repertoire of chemical processes available to primitive metabolic systems.

Box 26-4 METHODS

The SELEX Method for Generating RNA Polymers with New Functions

SELEX (systematic evolution of ligands by exponential enrichment) is used to generate aptamers, oligonucleotides selected to tightly bind a specific molecular target. The process is generally automated to allow rapid identification of one or more aptamers with the desired binding specificity.

Figure 1 illustrates how SELEX is used to select an RNA species that binds tightly to ATP. In step , a random mixture of RNA polymers is subjected to “unnatural selection” by passing it through a resin to which ATP is attached. The practical limit for the complexity of an RNA mixture is about $10^{15}$ $10 Superscript 15$ different sequences, which allows complete randomization of 25 nucleotides ( $4^{25} = 10^{15}$ $4 Superscript 25 Baseline equals 10 Superscript 15$ ). For longer RNAs, the RNA pool used to initiate the search does not include all possible sequences. RNA polymers that pass through the column are discarded (step ); those that bind to ATP are washed from the column with salt solution and collected (step ). In step , the collected RNA polymers are amplified by reverse transcriptase to make many DNA complements to the selected RNAs, then an RNA polymerase makes many RNA complements of the resulting DNA molecules. Finally, in step , this new pool of RNA is subjected to the same selection procedure, and the cycle is repeated a dozen or more times. At the end, only a few aptamers — in this case, RNA sequences with considerable affinity for ATP — remain.

FIGURE 1 The SELEX procedure.

Critical sequence features of an RNA aptamer that binds ATP are shown in Figure 2; molecules with this general structure bind ATP (and other adenosine nucleotides) with $K_{d} < 50 μ M$ $upper K Subscript d Baseline less-than 50 mu upper M$ . Figure 3 presents the three-dimensional structure of a 36 nucleotide RNA aptamer (shown as a complex with AMP) generated by SELEX. This RNA has the backbone structure shown in Figure 2.

FIGURE 2 RNA aptamer that binds ATP. The shaded nucleotides are those required for the binding activity.

FIGURE 3 RNA aptamer bound to AMP. The bases of the conserved nucleotides (forming the binding pocket) are pale green; the bound AMP is red. [Data from PDB ID 1RAW, T. Dieckmann et al., RNA 2:628, 1996.]

The figure shows a horizontal structure that has a wider, light green center, a triangular dark green piece to the lower left and right, and a yellow piece to the far right. The 3 prime end of a green strand begins at the lower left and curves cross the bottom, past a labeled G at the bottom of the light green region and beneath red A M P, to continue through the dark green part to the right, loop up and around in a yellow right-hand loop, and then curve back across a loop in the light green top center before folding back down to tend at 5 prime to the right of the original 3 prime starting point. A bracket over the light green region reads, G G A A G A A A C U G. A M P is shown in the center of the light green region as a wireframe model with the light green loop extending beneath it and then back up.

In addition to its use in exploring the potential functionality of RNA, SELEX has an important practical side in identifying short RNAs with pharmaceutical uses. Finding an aptamer that binds specifically to every potential therapeutic target may be impossible, but the capacity of SELEX to rapidly select and amplify a specific oligonucleotide sequence from a highly complex pool of sequences makes this a promising approach for generating new therapies. For example, one could select an RNA that binds tightly to a receptor protein prominent in the plasma membrane of cells in a particular cancerous tumor. Blocking the activity of the receptor, or targeting a toxin to the tumor cells by attaching it to the aptamer, would kill the cells. SELEX also has been used to select DNA aptamers that detect anthrax spores. Many other promising applications are under development.

4. The Structure of the Ribosome

As we shall see in Chapter 27, some natural RNA molecules, components of ribosomes, catalyze the formation of peptide bonds, offering a glimpse of how the RNA world might have been transformed by the greater catalytic potential of proteins. The evolution of a capacity to synthesize proteins would have been a major event in the RNA world, allowing the generation of polymers that could greatly stabilize complex RNA structures. However, the onset of peptide synthesis would also have hastened the demise of the RNA world. Proteins simply have more catalytic potential. The information-carrying role of RNA may have passed to DNA because DNA is chemically more stable. RNA replicase and reverse transcriptase may be modern versions of enzymes that once played important roles in making the transition to the modern DNA-based system.

5. Extant Vestiges of an RNA World

The known functions of RNA continue to multiply with each decade. Retroviruses, other RNA viruses, and retrotransposons inhabit a semi-independent universe, maintaining a parasitic existence within the biosphere. For evolutionary biologists, these almost-living entities provide a window on key steps in the evolution of life. Transposons may have been an early innovation in an RNA world. With the appearance of the first, inefficient self-replicators, transposition could have been an important alternative to replication as a strategy for successful reproduction and survival. Early parasitic RNAs would simply hop into a self-replicating molecule via catalyzed transesterification, then passively undergo replication. Natural selection would have driven transposition to become site-specific, targeting sequences that did not interfere with the catalytic activities of the host RNA. Replicators and RNA transposons could have existed in a primitive symbiotic relationship, each contributing to the evolution of the other. Modern introns, retroviruses, and transposons may all be vestiges of a “piggyback” strategy pursued by early parasitic RNAs. These elements continue to make major contributions to the evolution of their hosts.

6. Progress in the Search for an RNA Replicator

The RNA world hypothesis requires a nucleotide polymer to reproduce itself. Can a ribozyme bring about its own synthesis in a template-directed manner? Researchers are getting closer to finding such a ribozyme or ribozyme system. For example, Gerald Joyce and colleagues, in 2009, reported on the first set of two ribozymes that could cross-catalyze each other’s formation (Fig. 26-41). One ribozyme, E, catalyzes the joining of two oligonucleotides ( $A^{'}$ $upper A prime$ and $B^{'}$ $upper B prime$ ) to create a second, complementary ribozyme called $E^{'}$ $upper E prime$ . $E^{'}$ $upper E prime$ could then catalyze the joining of two other oligonucleotides (A and B) to form another molecule of E. In this system, the formation of E and $E^{'}$ $upper E prime$ was templated, and the amounts grew exponentially as long as substrates were available and proteins were absent. The system evolved so that more-efficient enzymes appeared in the population. A more general RNA-polymerase-like ribozyme was described in 2011 by Philipp Holliger and colleagues.

A two-part figure shows the self-sustained replication of an R N A enzyme with part a showing a possible reaction scheme and part b showing the ligation reaction involved. — FIGURE 26-41 Self-sustained replication of an RNA enzyme. This system has many of the properties of a living system. The RNA molecules incorporate information and catalytic function, and the reactions produce an exponential increase in the product RNAs. When variants of the RNA substrates are introduced, the system undergoes natural selection such that the best replicators eventually dominate the population. (a) A possible reaction scheme. Oligoribonucleotides A and B anneal to ribozyme $E^{'}$ $upper E prime$ and are ligated catalytically to form ribozyme E. The joining of oligoribonucleotides $A^{'}$ $upper A prime$ and $B^{'}$ $upper B prime$ is similarly catalyzed by ribozyme E. The levels of E and $E^{'}$ $upper E prime$ grow exponentially, with a doubling time of about one hour at $42 ° C$ $42 degree upper C$ , as long as there is a supply of the precursors A, B, $A^{'}$ $upper A prime$ , and $B^{'}$ $upper B prime$ . (b) The ligation reaction involves attack of the $3^{'}$ $3 prime$ OH of one oligoribonucleotide on the α phosphate of the $5^{'}$ $5 prime$ -triphosphate of the other oligoribonucleotide. Pyrophosphate is released. Base pairing of the substrates with the ribozyme plays a key role in aligning the substrates for the reaction. [Information from T. A. Lincoln and G. F. Joyce, *Science* 323:1229, 2009.]

FIGURE 26-41 Self-sustained replication of an RNA enzyme. This system has many of the properties of a living system. The RNA molecules incorporate information and catalytic function, and the reactions produce an exponential increase in the product RNAs. When variants of the RNA substrates are introduced, the system undergoes natural selection such that the best replicators eventually dominate the population. (a) A possible reaction scheme. Oligoribonucleotides A and B anneal to ribozyme $E^{'}$ $upper E prime$ and are ligated catalytically to form ribozyme E. The joining of oligoribonucleotides $A^{'}$ $upper A prime$ and $B^{'}$ $upper B prime$ is similarly catalyzed by ribozyme E. The levels of E and $E^{'}$ $upper E prime$ grow exponentially, with a doubling time of about one hour at $42 ° C$ $42 degree upper C$ , as long as there is a supply of the precursors A, B, $A^{'}$ $upper A prime$ , and $B^{'}$ $upper B prime$ . (b) The ligation reaction involves attack of the $3^{'}$ $3 prime$ OH of one oligoribonucleotide on the α phosphate of the $5^{'}$ $5 prime$ -triphosphate of the other oligoribonucleotide. Pyrophosphate is released. Base pairing of the substrates with the ribozyme plays a key role in aligning the substrates for the reaction. [Information from T. A. Lincoln and G. F. Joyce, *Science* 323:1229, 2009.]

In the center of part a, a molecule is shown that has a blue top strand labeled E and an orange bottom strand labeled E prime. The top strand has a long vertical protrusion upward on the left, the right strand has a long vertical protrusion downward on the right, and there are two small loops in the main strand. An arrow points counterclockwise to the three o’clock position accompanied by a reverse arrow indicating a reversible reaction. This yields E prime only. An arrow points counterclockwise to a structure at the ten o’clock position accompanied by a reverse arrow indicating a reversible reaction. The forward reaction is accompanied by a curved line showing the addition of a structure resembling E divided into two parts with a green piece labeled A on the left and a separate yellow piece labeled B on the right. The reverse reaction is accompanied by a branched arrow showing the loss of the same structure. This reaction yields E prime beneath A and B. An arrow points from this to the starting structure of blue E above orange E prime. A short clockwise arrow points from this structure to E by itself at the three o’clock position of a second cycle of arrows below. A reverse arrow is also present. Another clockwise arrow points from this position to the nine o’clock position, accompanied by a short line showing the addition of a structure resembling E prime divided into two halves, which are purple B prime on the left and red A prime on the right. The reverse reaction has a branched arrow showing the loss of these structures. This yields a structure with E above B prime to the left and A prime to the right. An arrow points from this back to the starting structure. All data are approximate. Part b shows a structure consisting of dark green A on the left, light green B on the upper right, and orange E prime below. There is a gap between U bonded to O H with a red pair of electrons on O at the end of the dark green A strand and p p p G at the beginning of the light green B strand. A red arrow points from the unpaired electrons to p p p G. The full structure is as follows. The top strand shows a green 5 prime strand labeled A that begins with unpaired G, then G above a dot, then paired U U C A U G U before bending upward with G bonded to C above bonded to U hydrogen bonded to A to the right below C hydrogen bonded to G to the right, then G with a dot to the right also bonded to A hydrogen bonded to U to the right and to U above with a dot to the right beneath U bonded to A to the right that begins a vertical string of hydrogen bonded nucleotides that continues as G U U A C before forming an unpaired top loop of G U A A and then returning down the right side with G U A A C A G with a dot to the left U, then a short loop to U to the right, then U with a dot to the left, then G bonded to C to the left, then A bonded to U to the left, then a slight loop of A U G before a new vertical strand begins of bonded G G U then U with a dot beneath, then a slight loop of G A A, then G with a dot beneath, then A A both hydrogen bonded below, then U with dot beneath bonded to O H above with a pair of red electrons on O. Light green B is to the right and begins with ppp G bonded to A below bonded to G with a dot beneath then to A C C all hydrogen bonded below with the final C bonded to G with a dot beneath bonded to C A A C U U all hydrogen bonded below before ending at unpaired A at the 3 prime end. From left to right, the orange bottom strand is 3 prime C U U A U with a dot above bonded A A G U A C G bonded to C C A G with a dot above, then a loop of A G bonded to U with a dot above, then U U both hydrogen bonded above, then G with a dot above, then a loop of A A G, then U with a dot above, then bonded U G G, then a bend vertically to G U A then bonded A G then U with a dot to the right, then U slightly to the left, then U bonded to the right, then G with a dot to the right, then A C A A G A U U C all bonded to the right before a loop of A A U G before the upward complementary strand of G A A U C U U G U U with a dot to the left then A bonded to the left and bonded to G above with e wdot to the left, then C and U bonded to the left, then C bonded to G above followed by U with a dot above connected to a bonded chain of G U U G A A ending in unbonded G G at the 5 prime end.

Although the RNA world remains a hypothesis, with many gaps yet to be explained, experimental evidence supports a growing list of its key elements. Further experimentation should increase our understanding. Important clues to the puzzle will be found in the workings of fundamental chemistry, in living cells, and perhaps on other planets.

SUMMARY 26.4 Catalytic RNA and the RNA World Hypothesis

Ribozymes and protein-based enzymes share common features, including folded three-dimensional structures, inactivation by denaturation, acceleration of reaction rates, saturable kinetics, and reaction specificity.
Ribozymes and RNA-based catalysts are present in organisms today and are involved in a wide range of activities, including tRNA processing, nuclear pre-mRNA splicing, and translation.
The evolution of life on Earth may have included an RNA world in which RNA was the central information carrier and catalyst before proteins and DNA emerged as key players. The existence of ribozymes provides a powerful piece of evidence in support of this hypothesis.