Skip to Main Content

Science & Technology

January 1, 2007
Volume 85, Number 01
p. 27-28

Aspirin's Dose Of Structural Insight

A recently identified crystal packing of aspirin reminds chemists that discoveries lurk in the most familiar places

Ivan Amato

Just over a year ago, researchers reported that they had isolated and characterized a long-sought second form of acetylsalicylic acid, known in most households as aspirin (J. Am. Chem. Soc. 2005, 127, 16802). Last April, lead author Michael Zaworotko of the University of South Florida, Matthew Peterson of TransForm Pharmaceuticals in Lexington, Mass., and several partners filed an international patent application, still pending, for the putative new form of aspirin.

Courtesy of Michael Zaworotko View Enlarged Image
Shades Of Difference In its long-known crystal form, Form I, aspirin molecules form dimers of dimers that align into layers, which tilt in alternating up-and-down directions. (The darker head of each arrow indicates that the underlying molecule is oriented up from the paper's plane.) In Form II, in alternating layers, the sets of dimers point toward or away from one another in upward- or downward-pointing V shapes.

The claim of a second crystal form—which highlights previously underappreciated subtleties in crystals and could have legal and commercial consequences in the aspirin industry—naturally caught the eye of Gautam R. Desiraju of the University of Hyderabad, in India. After all, Desiraju has long investigated, wrestled with, and written about fine structural subtleties and ambiguities in crystals, particularly crystal polymorphisms in which, as one common definition puts it, "the same substance exhibits different crystal packing arrangements." What's more, Desiraju and Zaworotko have known each other for years, having traveled in the same professional circles.

That's partly why Zaworotko was so taken aback by a pair of papers authored by Desiraju, Roland Boese of the University of Duisburg-Essen, in Germany, and Andrew Bond of the University of Southern Denmark and published online in late November in Angewandte Chemie. In prose replete with exclamation points and with a tone that conjures up that of a scolding headmaster, the first paper purports to deconstruct the claim of the Journal of the American Chemical Society communication with charges of shoddy crystallography and flawed data interpretation (Angew. Chem. Int. Ed., DOI: 10.1002/anie.200602378).

"It was such an attacking sort of paper," says Peterson, a crystallographer. The paper suggests that the data in the JACS communication could be just as easily interpreted as describing two orientations of the same already-known crystal form of aspirin as evidence for a new crystal form. In light of such purported ambiguity, Desiraju's team contends, the data reported in the JACS communication is insufficient for claiming to have found an elusive new crystal structure for aspirin.

The second Angewandte paper renders the first one all the more puzzling to Zaworotko and Peterson, because it confirms the existence of a second crystal form of aspirin (Angew. Chem. Int. Ed., DOI: 10.1002/anie.200603373). They concede that the crystallographic data reported in their JACS communication is not the best, but they argue that the validity of the resulting structure is reinforced by its conformity with an aspirin polymorph structure that theorist Sally Price of University College London proposed in 2004. Zaworotko notes his team obtained additional powder diffraction data that bolsters their case. Ironically, they weren't even going after an aspirin polymorph to begin with. The intriguing Form II came to light during cocrystallization experiments involving aspirin and the antiseizure compound levetiracetam.

What everyone agrees on is that the entire sequence of reports on polymorphism in aspirin puts the spotlight on fascinating knowledge gaps in organic chemists' understanding of the solid state. They say the collective message of the work is that aspirin molecules, and quite likely many other organic molecules, can assume subtly different packing arrangements even within the same minuscule crystal grain. Fundamental as that might sound, the identification and control of different crystal polymorphs is an important avenue in pharmaceutical development. Different polymorphs can have manufacturing advantages or distinct biomedical properties due to, say, variations in solubilities.

"Polymorphism is a huge driving force in pharma," comments Jerry Atwood, a supramolecular structure researcher at the University of Missouri. Thus, any claim of finding a new, possibly patentable polymorph in a commercially significant drug can carry high stakes, he says.

First synthesized in 1853, aspirin became a global best-selling medicine by the turn of the 19th century. In 2005, aspirin sales for Leverkusen, Germany-based Bayer alone amounted to roughly $800 million. Until the putative new polymorph was divulged, the entire aspirin industry appeared to have been built on a single crystal form.

"The distinction between the proposed aspirin polymorphs is subtle," write Desiraju, Boese, and Bond in their second Angewandte paper. So nuanced is the distinction that it takes some staring at the structural illustrations in the papers to discern the difference between the two forms. This subtlety also explains why it has taken so long to identify and detect the new aspirin crystal structure, researchers say. Both forms are composed of identical hydrogen-bonded dimers that lie in identical layers. The difference between the forms resides in how these layers stack. In Form I, sets of dimers are aligned, but adjacent layers tilt in opposite directions. In Form II, sets of dimers in each layer form subtle V shapes, which alternately point up and down in adjacent layers, resulting in a herringbone-like motif.

In their second paper, Desiraju's group concludes that the two forms coexist as "intergrowths" in crystalline aspirin and that the two structural types can toggle frequently and sharply throughout a crystal. Relevant to claims of polymorphs and patentability, Desiraju says, is that his group has not been able to make aspirin crystals containing only Form II; they always get some mix of the forms. This result calls into question just what should be considered a polymorph. "This even brings up the issue of what is a crystal," Desiraju says. Zaworotko counters that there is not enough experimental basis for concluding that synthesizing pure Form II is impossible.

Regarding the tone of his papers, Desiraju says, it "was intended to be provocative." He fears that quality control in crystallography is waning as new machines and software make this tricky analytic process—and the interpretation of the resulting data—seem easier and more accessible, even to nonspecialists, than it is. "I hate the word 'easy,' " Desiraju says.

In the first Angewandte paper, Desiraju, Boese, and Bond point out several shortcomings in the data that Zaworotko and his coauthors relied on in their JACS communication. For example, Desiraju's group points to the large differences between the actual crystallographic data reported in the communication and Price's theoretical aspirin polymorph. This discrepancy required Zaworotko's group to use what Desiraju and his colleagues deem to be an unacceptable amount of "refining." That's the process by which chemists derive an atomically detailed structure of a crystal's unit cell from the raw crystallographic data by way of an electron density map of the unit cell. Desiraju and his colleagues contend that their own new data provide a more defensible foundation for claiming the existence of Form II. They would not say whether they also were filing a patent application or a challenge to Zaworotko's.

No one contacted by C&EN expects the new discovery to have practical or commercial bearing on the aspirin industry, though Zaworotko notes that "if a gift horse comes my way, I won't turn it away." If it turns out that all aspirin manufacturing is producing some Form II that would be detectable with new quality-control protocols, for example, the patent holders of this composition of matter could conceivably claim royalty or license fees.

The larger payoff of the work, researchers say, is that it will alert pharmaceutical investigators to previously overlooked subtleties in the crystals of their medicinal compounds. Whether these end up being called polymorphs, intergrowths, polytypes, or something else, these structural details could well become central to high-stakes intellectual property claims.

Price, who says she is pleased that her theoretical proposals have inspired such intriguing and potentially valuable experimental studies, summed the general sentiment among those contacted by C&EN. "I see the new paper by Bond et al. as an exciting, more detailed refinement of the original discovery of Form II, illustrating how more extensive experimental work is revealing greater complexity in the organic solid state," Price says. "This is illuminating the gray area between natural variations in crystals."

Chemical & Engineering News
ISSN 0009-2347
Copyright © 2011 American Chemical Society