Friday, April 25, 2014

ASBMB Core Concepts in Biochemistry and Molecular Biology: Molecular Structure and Function

Theme

Better Biochemistry
The American Society for Biochemistry and Molecular Biology (ASBMB) has decided that the best way to teach undergraduate biochemistry is to concentrate on fundamental principles rather than facts and details. This is an admirable goal—one that I strongly support.

Over the past few months, I've been discussing the core concepts proposed by Tansey et al. (2013) [see Fundamental Concepts in Biochemistry and Molecular Biology]. The five concepts are:
  1. evolution [ASBMB Core Concepts in Biochemistry and Molecular Biology: Evolution ]
  2. matter and energy transformation [ASBMB Core Concepts in Biochemistry and Molecular Biology: Matter and Energy Transformation]
  3. homeostasis [ASBMB Core Concepts in Biochemistry and Molecular Biology: Homeostasis]
  4. biological information [ASBMB Core Concepts in Biochemistry and Molecular Biology: Biological Information]
  5. macromolecular structure and function [ASBMB Core Concepts in Biochemistry and Molecular Biology: Molecular Structure and Function]
This is the last of the core concept topics. It covers "Structure and Function," a theme that's been around for many decades. The Tansey et al. paper begins with ...
Biological Macromolecules are Large and Complex

Macromolecules are polymers of basic molecular subunits. They include the proteins (made up of amino acids), nucleic acids (comprised of nucleotides), carbohydrates (polymers of sugars), and lipids (with a variety of hydrophobic molecular constituents). The understanding of protein and nucleic acid structures is the most advanced. The functions of macromolecules include catalysis (protein and RNA enzymes), ligand-binding, cellular structure, signaling, and transport. Macromolecules can be very large. Nucleic acids in chromosomes have molecular weights ranging into the billions. Proteins (single polypeptide chains) range in size up to molecular weights of a million or more.

Students should be able to discuss the diversity and complexity of various biologically relevant macromolecules and macromolecular assemblies in terms of the basic repeating units of the polymer and the types of linkages between them.
This is the "structure" part of the topic. I would add a unit on stereochemistry—it's important for understanding amino acids and carbohydrates.

In discussing the diversity of proteins, it's important to talk about their evolutionary relationships. This is complicated but it's worth mentioning that you can group various proteins (or domains) into large structurally-similar families. Some of these might be related by descent and other might be examples of convergence. It's not always clear whether structural similarity indicates homology. (This is one of the reasons why "Homology Modelling" is a bad substitute for "Similarity Modelling" or "Comparative Moddling.")
Structure is Determined by Several Factors

Under a given set of conditions, most macromolecules tend to fold into one or a few semistable three-dimensional conformations. The structure of macromolecules is dictated by the sequence of amino acids (linked by peptide bonds), nucleotides (linked by phosphate diester bonds), or other constituents in the polymer, other covalent bonds linking the polymeric constituents, the pattern of weak noncovalent interactions between chemical groups within the folded structure, and the noncovalent and thermodynamic interactions between chemical groups of the structure and compounds present in its immediate environment.

Weak, noncovalent interactions—the hydrophobic effect, hydrogen bonds, ionic interactions, Van der Waal's interactions—play a special role in the folding of a large polymeric macromolecule into a particular active conformation. The hydrophobic effect makes the largest contribution to the structural stability of proteins and nucleic acids that are soluble in aqueous environments.

Portions of a macromolecule may fold up into a semistable conformation independently of other segments of the same macromolecule. These independently stable segments are called domains.

Students should be able to discuss the chemical and physical relationships between sequence and structure of macromolecules and evaluate chemical and energetic contributions to the appropriate levels of structure of the macromolecule and predict the effects of specific alterations of structure on the dynamic properties of the molecule.
This is mostly about protein folding—an important topic in biochemistry. The important concept is that polypeptides fold into a specific three-dimensional shape that corresponds to the lowest free energy form of the molecule. They need to know that the complex three-dimensional structure is made possible by rotation around the bonds in the polypeptide backbone but that the peptide bond restricts some conformations.

This is a good place to ram home the distinction between thermodynamics and kinetics. Although most proteins can fold in a reasonable amount of time, others might take a long time to fold. The folding reaction can be accelerated by molecular chaperones—very ancient proteins that are essential for survival.

Students need to understand the basic idea that folding proceeds through secondary structure intemediates such as β-stands, α-helices, and various turn motifs. They should be told that the final three-dimensional structure depends exclusively on the primary sequence of amino acids in some cases but there are additional requirements for many proteins. They need to know about the weak forces that cause and stabilize protein folding.
Structure and Function are Related

The function of a protein, nucleic acid, or other macromolecule is defined to a large extent by the specific molecular interactions it takes part in. Those interactions are in turn dictated by the structure of the macromolecule. If a protein or nucleic acid has a ligand-binding function, it will have a depression in its surface (a binding site) that is complementary in size, shape, charge, and other features to the ligand molecule to which it binds. If the protein or nucleic acid has a catalytic function, it will similarly have a depression on its surface (an active site) structured to facilitate catalysis of a particular reaction. If a protein or nucleic acid has a structural function, it will have a structure that confers strength or elasticity or whatever is required for a particular structural role in a cell or organism.

Students should be able to examine a structure of a macromolecule-ligand complex and predict the determinants of specificity and affinity and design experiments to test their hypothesis, explaining the basis of the proposed experiments and discussing potential results in the context of the hypothesis.
This is a pretty good beginning, although I would not specify that all active sites are in a depression on the surface. That's just not true.

I'm not a fan of getting students to "design experiments" to test hypotheses. That would require teaching a large number of techniques and there's no time for that in a concept-driven curriculum. It's hard enough to teach graduate students how to design experiments.
Macromolecular Interactions

The interactions between macromolecules and other molecules rely on the same weak, noncovalent interactions that play the major role in stabilizing the three-dimensional structures of the macromolecules themselves. The hydrophobic effect, ionic interactions, and hydrogen bonding interactions are prominent. The structural organization of interacting chemical groups in a binding site or an active site lends a high degree of specificity to these interactions.
This is the place where we need to teach the basic fundamentals of ligand binding. That includes the thermodynamics and kinetics of protein-ligand interactions. Students should know about dissociation constants and they should have a feel for the values that are functional inside a cell. This is an important concept for understanding allostery, regulation, and DNA-protein interactions.
Macromolecular Structures are Dynamic

Macromolecular structures are not static. Conformational changes large and small are often critical to function. Small changes can come in the form of localized molecular vibrations that can facilitate the access of small molecules to interior portions of the macromolecule. Large conformational changes can come in the form of the motions of different macromolecular domains relative to each other to facilitate catalysis or other forms of work.
The basic concept here is that proteins are not crystals in solution and that flexibility is an intrinsic part of function. Students need to know that "induced fit," for example, is not a characteristic of just some enzymes but that almost all enzymes exhibit some form of induced fit on binding substrates.

It's very difficult to get this concept across since so much of the learning material is two-dimensional and static (even when presenting three-dimensional objects).
Some Macromolecules are Intrinsically Unstructured

Segments of some proteins, and in a few cases entire proteins, are intrinsically unstructured. The unstructured segments often take up a particular three-dimensional structure when they interact with another macromolecule. The lack of structure in solution may facilitate a function in which interactions must occur promiscuously with several other molecules, as documented for some proteins with a signaling function.
I'm not sure this is a basic concept. I would teach it, but I'm not sure students would suffer is they didn't know this.
Macromolecular Function is Subject to Regulation

Following completion of polypeptide synthesis by the ribosome (post-translation), nascent polypeptides are almost invariably covalently modified in some functionally important way before, during, and/or after folding into their three-dimensional conformations. A wide variety of possible covalent modifications (e.g. partial proteolytic cleavage, intrachain and/or interchain disulfide formation, glycosylation, and phosphorylation) occur, and play a role in regulation, cellular targeting of the protein, or directly in the protein's function. Nucleic acids also undergo a variety of modifications that affect their function, including the modification of bases in RNAs, and the methylation of some cytosine residues in eukaryotic DNA.

Students should be able to compare and contrast the potential ways in which the function of a macromolecule might be affected and be able to discuss examples of allosteric regulation, covalent regulation, and gene level alterations of macromolecular structure/function.
This is far too complicated to discuss here. I really don't know how much regulation falls under the category of required basic concepts and how much is fluff on top of an already complicated subject. For example, one or two good examples of signal transduction pathways is probably enough but many teachers want to cover a dozen or more.

A few well-understood examples of allostery are enough if they are explained properly. You don't need to describe the details of allosteric regulation for every pathway. It's becoming clear to me that hemoglobin is NOT the best example and we should stop teaching it.

Similarly, one or two examples of covalent modification are sufficient if the concept is taught properly. Pick the best examples. Teach about chymotrypsin and forget about any other examples of proteolytic cleavage. I'm sure students will get the point.


Tansey, J.T., Baird, T., Cox, M.M., Fox, K.M., Knight, J., Sears, D. and Bell, E. (2013) Foundational concepts and underlying theories for majors in “biochemistry and molecular biology”. Biochem. Mol. Biol. Educ., 41:289–296. [doi: 10.1002/bmb.20727]

10 comments:

  1. Hi Larry

    re: It's becoming clear to me that hemoglobin is NOT the best example and we should stop teaching it.

    Oh dear... I hate it when you do that. I do exactly that - I employ hemoglobin as a paradigm when discussing the phenomenon of cooperativity in biology. I offer some snippets from my notes below. I woud welcome feedback or correction (from any and all) on the understanding the target audience be high school students.


    Consider Hemoglobin’s binding to Oxygen or to Carbon Dioxide. The protein’s “butterfingers” grip on the ligand is somewhat tenuous and slippery so the ligand can easily slip off. The ligand will only stay bound when driven by a so-called concentration effect together with a "cooperativity effect". If surrounding oxygen is in high concentration (and surrounding Carbon Dioxide is in low concentration) hemoglobin preferably binds oxygen (and vice versa). Introduce Carbon Monoxide, and all bets are off. The protein Hemoglobin irreversibly binds Carbon Monoxide and death quickly ensues. Reversibility (and cooperativity) of binding is an important concept in gene regulation and in Biology in general.

    …the quaternary structure of Hemoglobin permits cooperative binding and release of Oxygen and Carbon Dioxide in a very reversible manner, as discussed above. Hemoglobin is more efficient than Myoglobin by its superior ability to release Oxygen even though it is less efficient at binding Oxygen. Similarly, DNA binding proteins exist either as dimers or tetramers that both bind and release DNA, again exhibiting the same kind of kinetics as Hemoglobin.

    ...But if DNA binding proteins exist as mirror-image dimers or tetramers, the DNA being bound must present sequences that are mirror image repeats, our so-called palindromes or inverted repeats. Hold that thought!

    ...The Long Terminal Repeats that flank transposons and retroviruses also contain (inverted repeats) like those found in the CAP protein. Long Terminal Repeats turn on transposon and retroviral genes as well as enhancing the expression neighboring host genes. The “enhancer story” starts with bacteria!

    ReplyDelete
  2. I'm not a fan of getting students to "design experiments" to test hypotheses. That would require teaching a large number of techniques and there's no time for that in a concept-driven curriculum. It's hard enough to teach graduate students how to design experiments.

    As a layperson I wonder: Could they be "thought experiments"? In other words: How well would students have to know the necessary lab techniques in order to design a conceptual test of which of two or more hypotheses was correct?

    ReplyDelete
    Replies
    1. http://www.pbs.org/parents/dinosaurtrain/activities/a-colorful-hypothesis/

      Delete
    2. Could they be "thought experiments"?
      To a certain degree. But in biology especially, experiments really are quite limited by techniques. There are many interesting biological questions that haven't been explored not because nobody has thought about them but because there isn't a good technique for addressing them.

      Delete
  3. I would very much like to hear the professors opinion on this article:
    http://www.icr.org/article/8017/

    ReplyDelete
    Replies
    1. Sorry, I don't have either the time nor the inclination to analyze an article on the Institute for Creation Research website claiming that the comparison of mitochondrial DNA sequences shows that humans, roundworms, fruit flies, and water fleas were created less than 10,000 years ago.

      Delete
    2. I'll save you the time. The "study" is incredibly stupid. Or shameless*. Or both.

      Method: (According to the paper) scientists say some flies evolved ~20,000,000 years old. Scientists have measured a mutation rate for fruit fly mitochondrial DNA. Multiple 20,000,000 by that annualized rate. Whola: 2.3 million predicted differences. Ignore that the whole damn fly mitochondrial genome is only 20,000 bp. Ignore the common ancestor of living Drosophila melanogaster is likely not 20,000,000 years old (particularly the population studied in labs since the middle of last century). Few than "predicted" changes means creation must be young. Ta-freaking-da. Repeat for other species. YEC.

      Drosophila complete mitochondrial genome:
      http://megasun.bch.umontreal.ca/ogmp/projects/other/genbank/DMU37541.txt

      * Dr. Jeanson is Deputy Director for Life Sciences Research at the Institute for Creation Research and received his Ph.D. in cell and developmental biology from Harvard University.

      How can someone with a Ph.D. in cell biology from Harvard doesn't see the flaws in this "study"?

      Delete
    3. Note the footnote 5:
      "This is for a subset of the mitochondrial DNA, the “D-loop,” the only region of the human mitochondrial DNA for which a mutation rate has been measured to appropriate statistical confidence."
      Which is a bit like going at the question "How quickly can the average human run 100m?" and then saying that most data sources only time them to the second, so you went with the last summer games track finals where runs were times to 100ths of seconds...
      It's well know that mutation rates on the D-Loop are higher than for the rest of mtDNA by orders of magnitude.
      Apart from that it's not clear at all what mitochondrial genomes were compared. Obviously you measure a time to the most recent common maternal ancestor between individuals with this method, so it's highly relevant who's sampled.

      Delete
  4. I can understand Larry’s frustration.

    I cannot tell if this article is terribly illucid, merely incoherent or just a parody.

    Terms are bandied without definition.

    As far as I can tell, this article is suffers from two misconceptions:

    Misconception #7: Different Lineage Ages for Modern Species
    Misconception #8: Backwards Time Axes

    Found here

    Larry’s reticence is well placed. I draw everybody’s attention to the inclusion of the Institute for Creation Research on this perspicacious website

    ReplyDelete