Amino acid analyses of type 3 chondrites Colony, Ornans, Chainpur, and Bishunpur

Authors


Corresponding author. E-mail: z.martins@imperial.ac.uk

Abstract

Abstract– The CO3s Colony and Ornans and LL3s Chainpur and Bishunpur were analyzed for the first time for amino acids using gas chromatography–mass spectrometry (GC-MS). Type 3 chondrites have relatively unaltered metamorphic and petrological histories. Chainpur was the most amino acid rich of the four type 3 chondrites with a total amino acid abundance of 3330 parts per billion (ppb). The other type 3 chondrites had total amino acid abundances that ranged from 660 to 1110 ppb. A d/l ratio of <0.7 for all proteic amino acids suggests at least some amino acid terrestrial contamination. However, a small fraction of indigenous extraterrestrial amino acids cannot be excluded because of the presence of the nonprotein amino acid α-aminoisobutyric acid (α-AIB), and unusually high relative abundances (to glycine) of β-alanine and γ-ABA. The comparisons between the free and total amino acid contents of the samples also indicate a low free/total amino acid ratio (ranging from about 1:4 in CO chondrites to about 1:50 in Chainpur), which indicate that amino acids are present mainly in the bound form and were made detectable after acid hydrolysis.

Introduction

Chondrites comprise undifferentiated materials unaffected by processes associated with planetary evolution. In light of the petrology, mineralogy, and thermoluminescence properties, these meteorites are categorized into petrologic types 3–1 indicating increasing aqueous alteration, and types 3–6 showing increasing thermal metamorphism and petrologic equilibration. For several chondrite classes, the type 3 chondrites have been further divided into 3.0–3.9, with 3.0 being the most primitive petrologic type. Chondrites can be subdivided into carbonaceous and ordinary classes. Carbonaceous chondrites have elemental compositions that are the most primitive amongst other chondrites, and are less depleted in volatile elements (Botta and Bada 2002; Pizzarello 2006; Weisberg et al. 2006). Carbonaceous chondrites generally have low petrologic types (1–3, although the CKs extend to type 6) reflecting various degrees of aqueous and/or thermal alteration on the meteorite parent body. Ordinary chondrites are the most common materials and contribute to more than 85% of observed falls (Weisberg et al. 2006). Ordinary chondrites present a wide range of petrologic types from type 3 to 6 that are mostly higher than seen in the carbonaceous chondrites. The less altered members, commonly referred as unequilibrated ordinary chondrites (UOCs), are of remarkable interest because they have been subjected to very little aqueous and thermal alteration.

Some authors propose that type 3 meteorites, which are present in both carbonaceous and ordinary chondrites, represent the most pristine materials (e.g., McSween 1979; Bonal et al. 2007). Hence, they could be expected to provide crucial information about the early solar system conditions because the evidence therein has not been erased by secondary processes as may be the case for meteorites from other petrologic types.

The early Earth was frequently impacted by meteorites. These materials may have provided a rich source of organic matter for incorporation into the primitive prebiotic chemical evolution (e.g., Chyba and Sagan 1992; Ehrenfreund et al. 2002; Sephton and Botta 2008). Carbonaceous chondrites are frequently studied for prebiotic materials. They represent a primitive class of meteorite which contain up to 2–5% by weight of carbon, most of which are organic in nature (Ehrenfreund et al. 2002; Sephton and Botta 2008).

Ordinary chondrites also contain around 0.20 wt% carbon. The L/LL3.1 Bishunpur and LL3.4 Chainpur chondrites, however, were determined to contain notably high concentrations of carbon, with total carbon abundances of 0.53 and 0.44 wt%, respectively (Moore and Lewis 1967). They have been shown to contain a considerable amount of extractable organic materials, which implies that carbonaceous chondrites are not the only meteorite class that hosts organic compounds (Hayes 1967; Van Schmus and Wood 1967), despite their involvement in most of the chemical analyses (e.g., Cronin and Pizzarello 1983; Cronin 1989; Ehrenfreund et al. 2001; Sephton and Gilmour 2001; Botta et al. 2002; Glavin et al. 2006; Martins et al. 2006, 2007a). For example, extraterrestrial amino acids have been identified in two ureilites and in an H5 chondrite fragment from the Almahata Sitta parent body (Glavin et al. 2010; Burton et al. 2011).

Organic matter has a wide distribution amongst different meteorite classes. Therefore, it is important to identify the connections between carbonaceous chondrites and ordinary chondrites to interpret the primary distribution of carbon. We assume that such a connection will be most evident by comparing carbonaceous chondrites and the most primitive representatives of the ordinary chondrite population of the same petrological type, namely the type 3 chondrites.

Some of the carbon in meteorites is present as amino acids. Amino acids are the building blocks of proteins and enzymes, and their chirality allows discrimination between terrestrial and extraterrestrial origins. Most of the amino acid studies have analyzed extensively the CM carbonaceous chondrites, particularly Murchison (for a review, see, e.g., Martins and Sephton 2009), and also a number of other non-CM carbonaceous chondrites including the CV, CR, and CI chondrites (e.g., Cronin and Moore 1971, 1976; Ehrenfreund et al. 2001; Botta et al. 2002; Martins et al. 2007; Pizzarello et al. 2008). However, only limited amino acid analyses have been performed for meteorites from other groups. Previous amino acid analyses of ordinary chondrites include the H4 Forest Vale (Zenobi et al. 1992), four LL5 chondrites (Botta et al. 2008), and two LL6s (Martins et al. 2007b). In this study, the CO3s Colony and Ornans and LL3s Chainpur and Bishunpur were analyzed for amino acids using gas chromatography-mass spectrometry (GC-MS). These meteorites were studied to determine the presence of amino acids indigenous to the meteorites, and to evaluate the extents of terrestrial contamination. These compounds have the potential to reveal processes that occurred in the early solar system and any subsequent contamination episode following their fall to Earth.

Experimental Procedures

Samples, Chemicals, and Materials

The meteorite samples (Colony, Ornans, Bishunpur, and Chainpur) were obtained from the Natural History Museum, London. Colony (CO3.0) was found in 1975 (United States of America) when caught in the tines of a cotton cultivator (Nininger and Westcott 1984). Petrologic and chemical data indicate that it is one of the least metamorphosed CO3 chondrites (Rubin et al. 1985; Bonal et al. 2007). Ornans (CO3.4) fell in France in 1868. Bishunpur (L/LL3.1) and Chainpur (LL3.4) fell in India in 1895 and 1907, respectively.

Illite bought from Fluka was analyzed as a procedural blank for this work. The procedural blank was heated to 500 °C for 3 h and treated using the same experimental steps as the samples. All tools, glassware, and ceramics were sterilized by baking at 500 °C for 3 h. d- and l-isovaline (>99% purity) and ammonium hydroxide (NH4OH) (28–30 wt%, puriss. p.a.) were purchased from Acros Organics. All other amino acid standards (>99% purity), pyrene in methyl chloride (200 μL mL−1, analytical standard), and high-performance liquid chromatography (HPLC)-grade water (≤0.0003% nonvolatile matter) were purchased from Sigma-Aldrich. AG 50W-X8 resin (100–200 mesh) was acquired from Bio-Rad. Sodium hydroxide (NaOH, >99% purity) and hydrochloric acid (HCl) (37%, ≤5 ppm extractable organic substances) were purchased from Boom. The HPLC-grade dichloromethane (DCM, 99.8+% purity) was bought from Fisher Scientific. The trifluoroacetic anhydride/isopropanol (TFAA-IPA) derivatization kit, including acetyl chloride, was obtained from Grace Davison Discovery Sciences.

Sample Extraction and Desalting Procedures

The amino acid extraction procedures employed in this study are based on the method used in Martins et al. (2011). The meteorites Bishunpur (0.5005 g), Ornans (1.5036 g), Chainpur (1.002 g), Colony (2.0036 g), and an illite blank (1.5144 g) were examined. The samples were powdered and transferred to individual Pyrex test tubes (20 × 150 mm) for hot-water extraction and 1 mL of HPLC water was added to each sample. The test tubes were then flame-sealed and heated to 100 °C for 24 h in a heating block. After cooling to room temperature, the test tubes were rinsed with HPLC water, cracked open, and centrifuged for 5 min. Exactly half of the water supernatant (500 μL) was transferred to small test tube (10 × 75 mm), dried under vacuum, flame-sealed in larger test tube (20 × 150 mm) containing 6 N HCl, and then subjected to acid vapor hydrolysis (conducted in air) for 3 h at 150 °C. This method is rapid, clean, and does not destroy organics nor lose volatiles (Tsugita et al. 1987; Keil and Kirchman 1991). After the hydrolysis procedure, the test tubes were rinsed with HPLC water, and then cracked open. The small test tubes were removed and dried under vacuum. Both hydrolyzed and nonhydrolyzed samples were then brought up in 3 mL of HPLC water and desalted on a cation exchange resin. Amino acids were eluted with 5 mL of 2 M NH4OH. The eluates were collected in small test tubes, which were then evaporated to dryness under vacuum.

TFAA-IPA Derivatization and GC-MS Analysis

Prior to GC-MS analysis, amino acids were derivatized according to the method described in Martins et al. (2007a). Esterification with isopropanol and acylation with TFAA was carried out. The hydrolyzed and nonhydrolyzed samples and blanks were separately placed in 1 mL conical vials. A quantity of 100 μL of acetylchloride:isopropanol mixture (30:70 v/v) was added to each of the samples. The vials were tightly capped and the samples were heated in a heating block set at 110 °C for 1 h. After that, the samples were cooled to room temperature and dried under a gentle stream of dry N2. A quantity of 100 μL of DCM was added to the sample and dried under N2. 100 μL of DCM and 150 μL of TFAA were added to the dried sample. The vials were capped tightly again and heated to 100 °C for 10 min. The samples were then cooled to room temperature and the excess reagent was removed under a slow stream of N2. Prior to injection the derivatized samples were dissolved in 65 μL of DCM and 10 μL of pyrene in DCM as an external standard. The samples were transferred into GC autosampler vials and analyzed by GC-MS.

One μl of sample was injected into the GC-MS in splitless mode. Separations were carried out with an Agilent 6890 series GC equipped with an Agilent 5973 inert mass selective detector (MSD). The separations of the d,l-amino acid enantiomers were achieved using a Chirasil-l-Val column (50 m × 0.25 mm ID × 0.16 μm film thickness) from Alltech. Helium was used as carrier gas and the flow rate was set at 1 mL min−1. The source and quadrupole temperatures were maintained at 230 and 150 °C, respectively, while the injection port and the MSD transfer line were both heated to 220 °C. Standard autotunes with perfluorotributylamine (PFTBA) and air/water checks were made on a daily basis. The oven program was set at an initial temperature of 65 °C and held for 5 min, then increased by 2 °C min−1 to 80 °C and then 1 °C min−1 to 100 °C, then increased by 2 °C min−1 to 200 °C and held for 10 min, and finally increased by 10 °C min−1 to 225 °C and held for 5 min. Total ion current chromatograms were acquired and analyzed with Agilent Technologies MSD ChemStation software. Amino acids present in the meteorite samples were identified by comparison of the retention time and mass fragmentation pattern with known amino acid standard mixtures. Amino acids in the meteorite extracts were quantified by peak area integration of the corresponding ion fragment, which were then converted to abundances by using calibration curves. Single ion traces of the amino acids that are used to identify and quantify the amino acid content of samples are shown in Table 1. The calibration curves were created by plotting the amino acid standard/external standard (pyrene) target ion peak area ratio versus the mass of amino acid standard injected into the column (g).

Table 1.   Summary of the average total amino acid abundances (in ppb by weight) in the 6 M HCl acid-hydrolyzed, and nonhydrolyzed hot-water extracts of meteorite samples analyzed by GC-MS.a
Amino acidSingle ion (m/z)bCO chondritesLL chondrites
ColonyOrnansChainpurBishunpur
FreecTotaldFreecTotaldFreecTotaldFreecTotald
  1. aQuantification of the amino acids included background-level correction using a serpentine blank. The associated errors are based on the standard deviation of the average value between six separate measurements (N) with a standard error, rounded to one decimal place. δx = σx × N−1/2.

  2. bAmino acids were identified by the representative mass fragmentation patterns and comparison to the retention time of the amino acid standards. The abundances of each of the amino acids were acquired by peak area integration of the corresponding ion fragment.

  3. c,dFree amino acid content represents the nonhydrolyzed portions, with amino acids of free-form only. Total amino acid content represents the hydrolyzed portions, with amino acids of free- and bounded-forms.

  4. n.d. = Not detected above the detection limit of the GC-MS (10 pg of amino acid in a sample)

α-AIB154n.d.0.6 ± 0.1n.d.n.d.n.d.1.2 ± 0.3n.d.n.d.
d-Alanine14010.3 ± 0.213.3 ± 0.314.1 ± 0.522.3 ± 0.97.6 ± 0.368.2 ± 1.8n.d.41.8 ± 0.8
l-Alanine14016.7 ± 0.722.9 ± 0.726.1 ± 1.535.8 ± 1.68.6 ± 0.4126.4 ± 1.4n.d.61.0 ± 1.0
d-Valine168n.d.3.0 ± 3.0n.d.32.2 ± 0.6n.d.53.4 ± 0.5n.d.76.4 ± 0.3
l-Valine16815.3 ± 0.723.6 ± 0.731.5 ± 1.886.6 ± 2.2n.d.154.4 ± 2.5n.d.116.5 ± 1.4
Glycine12654.4 ± 2.0103.1 ± 2.8140.0 ± 4.4256.7 ± 7.253.0 ± 0.7534.6 ± 12.7n.d.294.2 ± 4.6
β-Alanine168/18534.7 ± 0.7121.6 ± 9.920.5 ± 4.415.5 ± 0.7n.d.340.1 ± 4.7n.d.232.5 ± 8.0
d-Leucine140n.d.n.d.n.d.n.d.n.d.4.4 ± 0.1n.d.n.d.
l-Leucine140n.d.n.d.n.d.n.d.n.d.34.3 ± 1.0n.d.15.1 ± 0.9
γ-ABA182n.d.112.1 ± 3.5n.d.n.d.n.d.655.2 ± 8.6n.d.n.d.
d-Aspartic acid184/2123.4 ± 0.15.4 ± 0.16.8 ± 0.29.7 ± 0.2n.d.45.9 ± 1.0n.d.22.4 ± 0.5
l-Aspartic acid184/2127.0 ± 0.312.0 ± 0.216.1 ± 0.315.5 ± 0.6n.d.69.0 ± 1.5n.d.32.8 ± 0.8
EACA210n.d.181.9 ± 2.4n.d.470.4 ± 13.0n.d.n.d.n.d.n.d.
d-Glutamic acid198/1806.8 ± 0.18.2 ± 0.28.8 ± 0.647.8 ± 1.6n.d.504.2 ± 2.918.2 ± 1.178.6 ± 3.9
l-Glutamic acid198/18042.0 ± 0.850.2 ± 0.746.9 ± 0.997.3 ± 2.5n.d.742.8 ± 6.748.2 ± 7.4140.6 ± 6.2
Total 1906603101090703330701110
Free:Total 1:31:41:481:16

Results and Discussion

The amino acid contents of the CO3s Colony and Ornans, and LL3s Chainpur and Bishunpur were analyzed using GC-MS. The single ion GC-MS traces (m/z 126, 140, 154, 168, 182, 184, 198, and 210) chromatograms of the derivatized (N-TFA, O-isopropyl) 6 N HCl-hydrolyzed and nonhydrolyzed meteorite hot-water extracts, procedural blank, and a mixture of amino acid standards are shown in Fig. 1. The chromatograms were scaled for clarity and the corresponding scale factors are shown in brackets in the figure. The amino acid concentrations in parts per billion (ppb) by weight of each of the samples are presented in Table 1 with the corresponding standard errors. Table 1 also shows the comparison between the “total” amino acid content (of the hydrolyzed fraction, composing amino acids of both free and bound forms) and the “free” amino acid content (of the nonhydrolyzed fraction, only free amino acids are present). The amino acid enantiomeric (d/l) ratios with the uncertainties are shown in Table 2. The uncertainties were calculated from the absolute errors of d- and l-amino acid abundances obtained by GC-MS, as shown in Table 1.

Figure 1.

 The 10–70 min region of the GC-MS chromatograms. Single ion GC-MS traces (m/z 126, 140, 154, 168, 182, 184, 198, and 210) of the derivatized (N-TFA, O-isopropyl) 6 N HCl-hydrolyzed meteorite hot-water extracts, illite blank, and amino acid standard. Similar chromatograms for the nonhydrolyzed fractions were obtained. The peaks were identified by comparing the retention time and mass fragmentation pattern to those in the amino acid standard run on the same day. d-Ala = d-Alanine; l-Ala = l-Alanine; d-Val = d-Valine; l-Val = l-Valine; Gly = Glycine; d-Norval = d-Norvaline; β-Ala = β-Alanine; l-Norval = l-Norvaline; d-Ser = d-Serine; l-Ser = l-Serine; d-Leu = d-Leucine; l-Leu = l-Leucine; d-Norleu = d-Norleucine; l-Norleu = l-Norleucine; d-Asp = d-Aspartic acid; l-Asp = l-Aspartic acid; d-Glu = d-Glutamic acid; l-Glu = l-Glutamic acid. The external standard Pyrene (m/z 202; eluted at 89 min) was added to each of the samples for amino acid quantification. The peak of pyrene did not co-elute with any amino acid and is not shown in the chromatographs.

Table 2.   Amino acid enantiomeric ratios (d/l) of the acid-hydrolyzed meteorite samples.a
Amino acidsCO chondritesLL chondrites
ColonyOrnansChainpurBishunpur
  1. aThe d/l ratios were obtained from the amino acid concentrations as listed in Table 1. The errors (uncertainties) are obtained by standard propagation calculations using the absolute errors shown in Table 1.

  2. n.d. = Not determined. Enantiomeric ratio cannot be calculated as the amino acid abundance falls below the detection limit.

Alanine0.58 ± 0.020.62 ± 0.040.54 ± 0.020.69 ± 0.02
Valine0.13 ± 0.130.37 ± 0.010.35 ± 0.010.66 ± 0.01
Leucinen.d.n.d.0.13 ± 0.004n.d.
Aspartic acid0.45 ± 0.010.63 ± 0.030.66 ± 0.020.68 ± 0.02
Glutamic acid0.16 ± 0.010.49 ± 0.020.68 ± 0.010.56 ± 0.04

Absolute Total Amino Acid Abundances

The total amino acid (free plus bound) abundances of the meteorite samples range from 660 ppb in Colony, 1090 ppb in Ornans, 1110 ppb in Bishunpur, to 3330 ppb in Chainpur. The CO3s Colony and Ornans have relatively low total amino acid concentrations (Table 1), much lower than the amino acid abundances of most other carbonaceous chondrites, which are generally at the parts per million (ppm) level (Martins and Sephton 2009; Glavin et al. 2011). Total amino acid abundances of Colony and Ornans are similar to CM1s SCO 06043 and MET 01070 (about 700 ppb, Glavin et al. 2011). The CR2 Graves Nunataks (GRA) 95229, for example, was shown to present the highest amino acid concentration for a meteorite of 249 ppm (Martins et al. 2007a), which is over 200 times the amino acid abundance of Ornans. The total amino acid abundances of the CO3 Colony and Ornans are only comparable to a few carbonaceous chondrites such as the CV3 Allende (1953 ppb) and the Antarctic meteorite CR1 Grosvenor Mountains (GRO) 95577 (900–1600 ppb) (Botta et al. 2002; Martins et al. 2007a; Glavin et al. 2011). CM and CO chondrites are in the same clan since they share similar chemical, mineralogical, and isotopic properties (Weisberg et al. 2006). However, the CM2 Murchison has a total amino acid abundance of 14,700 ppb (Martins et al. 2007b), which is notably higher than both the total amino acid concentrations of Colony and Ornans.

For ordinary chondrites, Martins et al. (2007b) has provided a set of information of L6 ordinary chondrites (Shişr 031 and Shişr 035). The LL3s Bishunpur and Chainpur demonstrate higher amino acid abundances than both Shişr 031 (60 ppb) and Shişr 035 (870 ppb). Shişr 031 and Shişr 035 have experienced extensive thermal alteration, which suggests that their amino acid contents represent terrestrial contamination. However, some extraterrestrial contribution may be present in LL3s Bishunpur and Chainpur (see d/l Enantiomeric Ratios and Terrestrial Contamination section). The high amino acid abundances of the LL3s in this research are noteworthy. Our data concur with the remarks from previous studies that the high carbon contents of type 3 ordinary chondrites (with regards to the amino acid abundances) are comparable to Type 3 carbonaceous chondrites (Moore and Lewis 1967; Van Schmus and Wood 1967).

Comparison Between the Free and Total Amino Acid Contents

Qualitative comparisons between the nonhydrolyzed (free) and hydrolyzed (total; free + bound) fractions of the meteorites analyzed in this study generally demonstrate increases in amino acid abundances, which illustrate that amino acids in these meteorites occur as both free amino acids and derivatives and/or acid-labile precursors that can be converted to free amino acids after acid-hydrolysis.

Considerable increases in amino acid concentrations were observed in all hydrolyzed hot-water sample extracts (Table 1). Total concentrations of free and total amino acids are 190 and 660 ppb, respectively, in Colony, 310 and 1090 ppb in Ornans, 70 and 3330 ppb in Chainpur, and 70 and 1110 ppb in Bishunpur. The total amino acid concentration in the LL3 Chainpur increased remarkably from 70 to 3330 ppb, representing an almost 50 times increase in amino acid content. The two CO3 chondrites show a threefold increase in total amino acid abundances after acid-hydrolysis. The remarkable increase in amino acid abundance after acid-hydrolysis is noteworthy in ordinary chondrites, which implies that the majority of amino acids in LL3s are present as bound amino acids.

Most of the amino acids (α-AIB, valine, leucine, γ-ABA, and EACA) being undetectable in the nonhydrolyzed fraction were observed at considerable concentration in the corresponding hydrolyzed fraction. For instance, l-glutamic acid and γ-ABA were not detected in the free form in Chainpur but were found at 743 and 655 ppb in the hydrolyzed portion. Although the significance of hydrolysis is less profound in the CO3 chondrites when compared with that of the LL3 samples, the amino acid concentrations of valine, glycine, γ-ABA, and EACA increased noticeably in both Colony and Ornans.

The ratios of free/total amino acids of the ordinary chondrites in this research are lower than many preceding meteorite studies (0.02 in Chainpur and 0.06 in Bishunpur), which shows that the petrologic type 3 meteorite extracts, particularly ordinary meteorites, contain a considerably larger amount of acid-labile precursors than free amino acids. The ratios of free to total amino acids were fairly similar in most of the moderately to heavily altered carbonaceous chondrites. The ratios are all within the range of 0.4 to 0.5 in CM1s Scott Glacier (SCO) 06043 and Meteorite Hills (MET) 01070, CM2s Murchison and Lewis Cliff (LEW) 90500, and CR1 GRO 95577 (Glavin et al. 2006, 2011; Botta et al. 2007). Nevertheless, the ratio is much lower in the CM2 Allan Hills (ALH) 83100 (0.08) due to the high concentration of the bound amino acid 6-aminocaproic acid (EACA) (Glavin et al. 2006).

Glavin et al. (2011) have observed lower ratios of free/total amino acids in the less altered chondrites. Their observations apparently contradict the assumption that free/total amino acid ratios are lower in the more aqueously altered chondrites due to the fact that bound amino acids are comparatively more resistant to oxidation during aqueous alteration. Our study supports the results presented by Glavin et al. (2011) because the less altered type 3 chondrites actually demonstrate lower ratios of free/total amino acids than chondrites of other petrologic types. The more aqueous alteration a chondrite has experienced, the more free amino acids will be released from their bound form, resulting in a higher ratio of free to total amino acids. In the same manner, the less altered type 3 chondrites should demonstrate lower free/total amino acid ratios. It is important to note that Glavin et al. (2011) studied CR3 chondrites and in this work we have studied CO3s and LL3s. Although they belong to the same petrologic type 3 chondrites, thermal histories are very different between CR3 (that escaped heating) and other chondrites. Our data seem to suggest that a low ratio of free/total in CO3s and LL3s is likely due to thermal metamorphism as free amino acids are lost by heating, rather than less aqueous alteration.

Individual Amino Acids Distributions

Various amino acids were found in the CO3s Colony and Ornans including α-aminoisobutyric acid (α-AIB), d,l-alanine, d,l-valine, glycine, β-alanine, γ-amino-n-butyric acid (γ-ABA), d,l-aspartic acid, EACA, and d,l-glutamic acid (Table 1). Amongst these amino acids, EACA, β-alanine, γ-ABA, and glycine are more abundant, ranging from 182 to 103 ppb in the hydrolyzed Colony (Table 1).

The LL3 chondrites Chainpur and Bishunpur contain all of the amino acids found in CO3 chondrites only without EACA, but they contain trace amounts of d,l-leucine which were not found in the CO3 samples. In the hydrolyzed LL3 Chainpur, l-glutamic acid was found to be the most abundant (743 ppb), followed by γ-ABA (655 ppb), glycine (535 ppb), d-glutamic acid (504 ppb), and β-alanine (340 ppb) (Table 1).

α-AIB and isovaline have rare occurrences in the terrestrial biosphere, and thus are characteristic of amino acids of extraterrestrial origin. In this study, α-AIB was detected in both Colony and Chainpur, but only present at a minute amount (≤1 ppb; Table 1). Its presence is unequivocal as determined by comparison of the retention time and mass spectrum to the standard. No isovaline was detected in the samples. Possible formation pathway for α-AIB includes the Strecker-cyanohydrin amino acid synthesis during aqueous alteration in the meteorite parent body. However, α-AIB detected in Colony and Chainpur could not been formed this way because the meteorite parent body experienced maximum temperatures of 500 °C (Keck and Sears 1987; Sears et al. 1991; Cody et al. 2008), causing rapid decomposition of amino acids (Rodante et al. 1992). More likely, α-AIB was formed from newly accreted amino acid precursor material, after the parent bodies of the Colony and Chainpur meteorites cooled to lower temperatures.

γ-ABA and β-alanine were also found at significant abundances in both the ordinary LL3 and carbonaceous CO3 chondrites. γ-ABA is present in Chainpur (about 655 ppb) as the second most abundant amino acid after l-glutamic acid and was found also in Colony (about 112 ppb). β-Alanine was detected in all acid-hydrolyzed samples at various concentrations (Colony: 122 ppb; Ornans: 15 ppb; Chainpur: 340 ppb; Bishunpur: 233 ppb). These two amino acids were also the two most abundant amino acids found in CI carbonaceous chondrites and were present at significant levels in the CM2 Murchison (Botta et al. 2002, 2007). The relative abundances of β-alanine and γ-ABA compared with glycine in the ordinary LL3 and carbonaceous CO3 chondrites analyzed in this study are very different than terrestrial sources (see discussion in the Relative Amino Acid Contents section). In addition, β-alanine present in the Orgueil meteorite was shown to be extraterrestrial in origin based on its carbon isotopic ratio (Ehrenfreund et al. 2001). The low amino acid abundances and the small mass obtained does not allow for the measurement of compound-specific isotope ratios. Based on the abundance of β-alanine and γ-ABA present in Chainpur and Colony (Table 1), a typical GC-C-IRMS limit of detection of a couple of nmols and the fact we needed at least three replicate isotope measurements of each amino acid, then we calculate that at least 2 and 6 g, respectively, of meteorite sample are needed. In the case of β-alanine present in Bishunpur and Ornans (no γ-ABA was detected above the detection limit of the GC-MS), we calculate that at least 3 and 35 g, respectively, of meteorite sample are needed. Without measuring the proper controls (soils from the landing site environments of the Colony, Ornans, Bishunpur, and Chainpur meteorites) and without carbon isotope measurements we cannot determine without a doubt the origin of the high levels of β-alanine and γ-ABA in the ordinary LL3 and carbonaceous CO3 chondrites analyzed in this study.

The EACA was found at high concentrations in the CO3 samples (182 ppb in Colony and 470 ppb in Ornans). A high concentration of bound EACA has also been found previously in the Antarctic meteorites such as the ALH 83100 (8480 ppb; Glavin et al. 2006). EACA is a typical contaminant contributed from nylon sample bags common to meteorite storage, and usually exists as a Nylon-6 peptide, detectable after acid-hydrolysis (Glavin et al. 2006). The relatively low levels of EACA in the CO3 and LL3 meteorites in comparison to Antarctic meteorites suggests that they may not have been exposed to nylon. Unfortunately, we do not know whether nylon was used to collect and store these meteorite samples. Therefore, we cannot conclude whether the EACA detected in these meteorites is indigenous.

A common unknown compound (compound X) is present in both hydrolyzed and nonhydrolyzed CO3 and LL3 samples at relatively high concentrations (Fig. 1). Since compound X was not detected in the illite blank, it is considered either a contamination obtained prior to laboratory handling (probably during sample curation or storage), or a compound indigenous to the CO3 and LL3 samples. The compound X eluted at 20 min. It contains a TFAA derivative in its structure (represented by the trifluoromethyl group of 69 m/z), which indicates the presence of an amine group. This unknown compound was determined at a significant amount in the nonhydrolyzed portion, which suggests its presence chiefly as free amino acid. The fragmentation pattern of compound X suggests a possible configuration of a CH2 carbon backbone at the α-position to the amine. The presence of the 126 and 139 m/z fragments also implies that the amine group is not bonded to the alpha-carbon (Cα), and that the amino acid is at least at a beta order as it is not glycine. Therefore, the compound X has a configuration that differs from α-AIB and isovaline without an alkyl group attached to the carbon at the α-position, and it should possess a structure similar to β-AIB, despite replacement of the alkyl group with all possible moieties. Compound Z is another unknown compound present at high abundance only in the hydrolyzed fraction of Colony (Fig. 1), i.e., present mainly in the bond form. As with compound X, compound Z was not detected in the illite blank. An amine group is represented by the trifluoromethyl fragment with m/z 69. The mass fragment m/z 196 is the most intense, followed by a gap of 28 amu to m/z 168 (cleavage of one C=O group), and sequentially by ions 14 amu apart (m/z 126, 140, 154) corresponding to cleavage of CH2 groups. This suggests that compound Z is a linear amino acid. It also suggests that the amine group is bonded to the delta-carbon, and that compound Z could possibly be 5-aminovaleric acid.

d/l Enantiomeric Ratios and Terrestrial Contamination

As discussed in the above section, identification of typical terrestrial and extraterrestrial amino acids in the samples allows determination of the extent of terrestrial contamination. Another technique that can aid in identifying the origins of the amino acids is the estimation of the amino acid d/l enantiomeric ratios for chiral amino acids. Homochirality of l-amino acids represents terrestrial contamination, usually demonstrated by d/l ratios <0.4. Otherwise, an equal mixture of both d,l-amino acids (reaching a racemic ratio, d/l = 1) generally indicates an abiotic synthetic origin. d/l ratios of 0.8–1.0, on the other hand, represent a significant proportion of indigenous extraterrestrial amino acid. d/l ratios that fall between 0.4–0.8 suggest an abiotic origin but a subsequent addition of terrestrial contaminant.

The d/l ratios of alanine, valine, leucine, aspartic acid, and glutamic acid of Colony, Ornans, Bishunpur, and Chainpur were calculated and presented in Table 2. The d/l ratios of the five amino acids of the meteorite samples all fall below 0.7, with the d/l glutamic acid ratio as low as 0.16 ± 0.01 in Colony and a d/l leucine ratio of 0.13 ± 0.00 in Chainpur (Table 2). Assuming that the enantiomeric ratios of indigenous proteic amino acids were racemic prior to the meteorite fall, l-enantiomeric excess (ee) would indicate terrestrial contamination obtained after the fall. Therefore, the low d/l ratios of the protein amino acids of the meteorites in this research indicate the addition of terrestrial l-amino acids during the meteorite residence time on Earth. In fact Colony has suffered considerable weathering (Rubin et al. 1985).

The samples in this study have d/l ratios between 0.49–0.63 for Ornans, 0.54–0.68 for Chainpur, and 0.56–0.69 for Bishunpur (Table 2). The degree of terrestrial contamination is dependent on (1) the terrestrial residence age (there is a correlation between contamination level and the length of terrestrial residence as seen by Bland et al. [1996] and Martins et al. [2007b]), (2) the petrographic characteristic of the sample (e.g., porosity, metal content), influencing the degree of absorption of contaminants, and (3) availability of contaminants on the meteorite fall site or sample curation/handling.

Amino acids detected in most of the ordinary chondrites in previous studies were shown to be derived from terrestrial contamination. The ordinary chondrite H4 Forest Vale was shown to contain predominantly glycine, l-aspartic acid, and l-alanine (Zenobi et al. 1992). d/l ratios of valine, aspartic acid, glutamic acid, and alanine of Shişr 035 were between 0.18 and 0.57 (Martins et al. 2007b). However, the d/l ratios of our samples, particularly the LL3s Chainpur and Bishunpur (d/l ratio of aspartic acid as high as 0.66 and 0.68, respectively) (Table 2), are notably higher than most of the contaminated ordinary chondrites, giving them a possible initial source of extraterrestrial components and subsequent terrestrial (l-amino acid) additions. It is important to note that the amino acid contents (both total and individually) in these meteorites seem to be correlated with the metamorphic grades (see the Relationships Between Petrologic Type and Organic Contents section) in the order Chainpur > Bishunpur > Ornans > Colony, suggesting that the detected amino acids are potentially derived from the meteorite parent bodies rather than terrestrial contaminants. The fact that the majority of amino acids in LL3s are present as bound amino acids (Table 1, and the Comparison Between the Free and Total Amino Acid Contents section) also supports the possibility that contamination is low. The most reliable analyses for determining the source of meteoritic amino acids is to measure their compound specific isotope ratios. As stated before, the low amino acid abundances of the meteorites do not allow performing these analyses and therefore, it is not possible to confirm if any of the amino acids are exclusively extraterrestrial. However, the contribution of terrestrial contaminant may be estimated by mass balance calculation of carbon isotope analyses and d/l ratios (Martins et al. 2007a). If we assume that the indigenous meteoritic material is racemic, that its d- and l-enantiomers have the same δ13C composition, and if we use δ13C values of +46.1‰ for d-glutamic acid and −19.5‰, for l-glutamic acid (Martins et al. 2007), the heaviest isotopic composition for the glutamic acid contaminant allowed by the abundance errors is δ13C = −32‰, −83‰, −158‰, and −103‰ for Colony, Ornans, Chainpur, and Bishunpur, respectively. Except for Colony, these δ13C values are outside the terrestrial range for amino acids, which range from −60.93‰ to −0.30‰ (Scott et al. 2006). δ13C values for aspartic acid, valine, and alanine present in the Colony, Ornans, Chainpur, and Bishunpur meteorites all fell outside the terrestrial range. These results as well as the racemic enantiomeric ratios (Table 2), suggest that although terrestrial contamination is possible, some of the glutamic acid, aspartic acid, valine, and alanine present in the meteorites are indigenous.

Relationships Between Petrologic Type and Organic Contents

The samples analyzed in this research are all petrologic type 3 chondrites. Type 3 chondrites have been suggested to contain the most pristine matter owing to their relatively unaltered metamorphic and petrological histories. There is a notable correlation between amino acid abundance and alteration extent of meteorites. For example, the Antarctic CR3 chondrite Queen Alexandra Range (QUE) 99177 is recently determined to be less altered than the other CR2 chondrites (Floss and Stadermann 2009), and the amino acids therein are well preserved, showing a striking amino acid abundance as high as 81,000 ppb (Glavin et al. 2011). Contrastingly, the LL6s Shişr 031 and Shişr 035 have undergone extensive thermal metamorphism at peak temperature of more than 800 °C (Slater-Reynolds and McSween 2005). The amino acids they contain may be largely derived from terrestrial contamination following their fall to Earth as temperatures greater than 500 °C cause rapid decomposition of amino acids (Rodante et al. 1992). However, an extraterrestrial contribution may not be discarded. Work by Glavin et al. (2010) and by Burton et al. (2011) revealed the presence of extraterrestrial amino acids in two different ureilites that had been exposed to at least 1200 °C on the parent body. They argued that the amino acids formed by Fischer-Tropsch/Haber-Bosch type gas-grain reactions after the parent body cooled to lower temperatures.

It is appropriate to consider the variations between amino acid contents of meteorites of the same petrologic type. The amino acid abundances of the LL3s in this research are shown to be higher than most of the petrologic type 3 counterparts. For instance, CV3 chondrites were reported to contain only low to trace levels of amino acids (Cronin and Moore 1971, 1976), while Mokoia was determined to be almost devoid of amino acids, with a total amino acid abundance of less than 1 nmol g−1 (about 100 ppb) (Cronin and Moore 1976). As we are aware, the only type 3 chondrite determined to possess the highest amino acid abundance is CR3 QUE 99177 (81,000 ppb) (Glavin et al. 2011).

In the light of the maturation of organic matter, CO3.4 Ornans is more metamorphosed than CO3.0 Colony, and the ordinary chondrite LL3.4 Chainpur is more thermally altered than LL3.1 Bishunpur (Quirico et al. 2003; Bonal et al. 2007). Although the organic content is generally expected to be lower for samples with higher metamorphic grades as organic materials are destroyed at elevated temperatures, this assumption does not seem to agree with the data obtained in this research. The thermally altered samples were shown to contain higher abundances of organic components. The total amino acid content is 1090 ppb for CO3.4 Ornans but only 190 ppb for CO3.0 Colony. A similar observation was made for ordinary chondrites, where the total amino acid abundance of LL3.4 Chainpur is 3330 and 1110 ppb for LL3.1 Bishunpur. Cody et al. (2008) quantitatively evaluated the peak temperature of Chainpur and Bishunpur based on the highly conjugated sp2 carbon structure of organic macromolecules. The temperature for Bishunpur (551 °C ± 34) is higher than that for Chainpur (483 °C ± 63). Although the petrologic orders are opposite in this case, the higher total amino acid abundance for Chainpur when compared with Bishunpur seem to be consistent with the study by Cody et al. (2008), suggesting that the amino acid contents are influenced by metamorphic temperature. In addition, Bishunpur was shown to have experienced hydrous alteration, which may be the reason for its lower amino acid abundance (Alexander et al. 1989). On the other hand, Sears et al. (1991) suggested that the CO chondrites have experienced longer periods of lower temperatures than the ordinary chondrites. According to our study, the ordinary chondrites, which have experienced higher temperature in their history, demonstrate higher organic contents than the carbonaceous chondrites. It is possible that amino acids may be formed from Fischer-Tropsch/Haber-Bosch type gas-grain reactions after the meteorite parent body cooled to much lower temperatures (Glavin et al. 2010; Burton et al. 2011). In addition, amino acids may be formed during the cooling process in the parent body still at elevated temperatures but lower than 500 °C (Rodante et al. 1992). The low organic content of the CO3.0 Colony can be explained by its weathering history. Colony has been badly weathered (Rubin et al. 1985); therefore, the organic materials it contains would have been oxidized or leached away during the weathering process.

Relative Amino Acid Contents

To demonstrate the correlation between amino acid content and the alteration extent of meteorites, the relative amino acid abundances of Colony and Ornans were compared with that for other carbonaceous chondrites obtained from previous studies (Table 3; Fig. 2), while that of Chainpur and Bishunpur were compared with the ordinary chondrites (Table 4; Fig. 3). The relative amino acid contents (glycine = 1) of β-alanine, d-alanine, and γ-amino-n-butyric acid (γ-ABA) in Colony, Ornans, Chainpur, and Bishunpur were calculated and presented in Tables 3 and 4. Note that the amount of terrestrial amino acid contamination is not well constrained for glycine in the CO3s and LL3s analyzed in this study, and the relative ratio could vary considerably between meteorites depending on amount of glycine contamination each sample was exposed to.

Table 3.   Relative amino acid abundances (glycine = 1) of carbonaceous chondrites.a
AA/glycineCR1 GRO 95577 [1] CM1
ALH 88045 [2]
CR2
GRA 95229 [1]
CM2
Murchison [2]
CR3
QUE 99177 [3]
CV3
Allende [4] 
CO3.0
Colonyb
CO3.4
Ornansb
  1. aRelative amino acid abundances calculated from the absolute amino acid concentrations of the acid-hydrolyzed hot-water extracts of the carbonaceous chondrites reported in Table 1 and also from previous analyses. In case only upper limits of the absolute concentrations were measured, upper limits for the relative concentrations were calculated.

  2. bThis study.

  3. n.d. = Not detected.

  4. References: [1] Martins et al. 2007a; [2] Botta et al. 2007; [3] Glavin et al. 2011; [4] Botta et al. 2002.

β-Alanine0.90 ± 0.100.53 ± 0.180.08 ± 0.010.65 ± 0.080.14 ± 0.040.69 ± 0.191.18 ± 0.100.06 ± 0.00
d-Alanine0.54 ± 0.171.05 ± 0.411.30 ± 0.170.33 ± 0.090.25 ± 0.07<0.010.13 ± 0.000.09 ± 0.00
γ-ABA0.40 ± 0.06 0.46 ± 0.17 0.05 ± 0.010.41 ± 0.07 0.09 ± 0.02 0.67 ± 0.33 1.09 ± 0.04 n.d.
Figure 2.

 A comparison of the relative amino acid abundances (glycine = 1) of β-alanine (dots), d-alanine (light downward diagonal), and γ-amino-n-butyric acid (wide upward diagonal) in the acid-hydrolyzed hot-water extract of carbonaceous chondrites (CR1 GRO 95577, CM1 ALH 88045, CR2 GRA 95229, CM2 Murchison, CR3 QUE 99177, CV3 Allende, CO3.0 Colony, and CO 3.4 Ornans) based on the data listed in Table 3.

Table 4.   Relative amino acid abundances (glycine=1) of ordinary chondrites.a
AA/glycineLL3.1
Bishunpur
LL3.4
Chainpur
LL5
LAP 03624b,[1]
L6
Shişr 035b,[2]
L6
Shişr 031b,[2]
  1. aRelative amino acid abundances calculated from the absolute amino acid concentrations reported in Table 1, measured in this study by GC-MS. In case only upper limits of the absolute concentrations were measured, upper limits for the relative concentrations were calculated.

  2. bThe data for the LL5 LAP 03624, L6s Shişr 031, and Shişr 035 were calculated from the acid-hydrolyzed hot-water extracts of the ordinary chondrites in previous analyses, measured by HPLC-FD.

  3. n.d. Not detected.

  4. References: [1] Botta et al. 2008; [2] Martins et al. 2007b.

β-alanine0.64 ± 0.020.79 ± 0.030.97 ± 0.340.20 ± 0.080.38 ± 0.19
d-alanine0.13 ± 0.000.14 ± 0.00n.d.0.68 ± 0.34<0.08
γ-ABA1.22 ± 0.03n.d.2.38 ± 0.78<0.05<0.08
Figure 3.

 A comparison of the relative amino acid abundances (glycine = 1) of β-alanine (dots), d-alanine (light downward diagonal), and γ-amino-n-butyric acid (wide upward diagonal) in the acid-hydrolyzed hot-water extract of ordinary chondrites (LL3.1 Bishunpur, LL3.4 Chainpur, LL5 LAP 03624, and L6s Shişr 031 and Shişr 035) based on the data listed in Table 4.

For carbonaceous chondrites, as shown in Fig. 2, there is a trend of increasing γ-ABA/glycine ratio but decreasing d-alanine/glycine ratio from the type 1 chondrites to the less altered type 3 chondrites. The γ-ABA/glycine ratio increased from 0.4 in CR1 GRO 95577 to 1.09 in Colony (Table 3) and the d-alanine/glycine ratio decreased from 0.9 in CM1 ALH 88045 to 0.13 in Colony. Glavin et al. (2011) described a positive correlation between the β-alanine/glycine ratio to aqueous alteration. However, this relationship is not obvious in our data (Fig. 2). Ornans is slightly metamorphosed and was shown to have experienced higher aqueous alteration than other CO3s (Bonal et al. 2007). Compared with other carbonaceous chondrites, as indicated by our data, Ornans contains substantially lower amino acid (β-alanine, d-alanine, and γ-ABA) abundances relative to glycine (Fig. 2). On the other hand, the relative amino acid distribution of Colony is similar to the CV3 Allende in that the ratios of β-alanine and γ-ABA to glycine are both very high (Fig. 2). As for ordinary chondrites, there is no obvious correlation between the relative amino acid distributions and thermal alteration (Fig. 3). The β-alanine and d-alanine abundances (relative to glycine) in Bishunpur (0.64 and 0.13, respectively) and Chainpur (0.79 and 0.14, respectively) are very similar (Table 4), while the γ-ABA abundance (relative to glycine) in Bishunpur (1.22) is very high.

It is possible to observe very similar patterns in amino acid relative abundances for the CV3 Allende, CO3 Colony, LL3 Bishunpur, and LL5 LAP 03624, which may provide information about indigenous amino acids correlated with thermal metamorphism. For instance, α-amino acids (including α-AIB and isovaline) could be preferentially formed via low temperature aqueous alteration reaction (such as in the CM and CR parent bodies), while β- and γ-amino acids could be preferentially formed via heat-related formation processes that type 3 chondrites experienced (see the Relationships Between Petrologic Type and Organic Contents section).

The high relative abundances of β-alanine and γ-ABA compared with glycine in the LL3s and CO3s (Tables 3 and 4) are unusual and very different from terrestrial sources. Amino acid analyses of terrestrial soils from several worldwide locations show relative abundances of β-alanine and γ-ABA of ≤0.2 and ≤0.1, respectively (e.g., Garry et al. 2006; Martins et al. 2007b, 2011). Therefore, we cannot exclude an extraterrestrial source for at least some of the β-alanine and γ-ABA present in the ordinary LL3 and carbonaceous CO3 chondrites analyzed in this study.

Summary and Conclusion

The free and total amino acid contents of the carbonaceous chondrites CO3s Colony and Ornans and ordinary chondrites LL3s Bishunpur and Chainpur were determined by GC-MS analysis. The samples were analyzed for the first time for amino acids. These COs and LLs are all petrologic type 3 chondrites. The total amino acid abundances (free plus bound) of the acid-hydrolyzed hot-water sample extracts are 660 ppb in Colony, 1090 ppb in Ornans, 1110 ppb in Bishunpur, and 3330 ppb in Chainpur. The amino acid contents of the CO3s are generally lower than most of the carbonaceous chondrites. However, the ordinary chondrites, in particular Chainpur, contain substantial amino acid contents that are the highest amongst other previously studied ordinary chondrites, and are shown to be comparable to some carbonaceous chondrites.

The ratios of free to total amino acids were obtained for the chondrites. This allows the determination of the proportion of acid-labile precursors present in the meteorites. The increase in amino acid concentrations was prominent. There was almost a 50 times increase in the total amino acid abundance in LL3 Chainpur after acid-hydrolysis. This result demonstrates that amino acids, notably in ordinary chondrites, occur chiefly as acid-labile precursors which are detectable after acid-hydrolysis.

A wide range of amino acids are present in the samples. The most abundant amino acids detected in the CO3s are EACA and glycine (182 and 103 ppb, respectively, in Colony, and 470 and 257 ppb in Ornans), and in the LL3s are l-glutamic acid and glycine (743 and 535 ppb in Chainpur, and 141 and 294 ppb in Bishunpur). Amino acids found at significant levels in both meteorite groups include β-alanine and γ-ABA. The high relative abundances of β-alanine and γ-ABA compared with glycine are unusual and very different from terrestrial sources, suggesting some fraction of indigenous extraterrestrial contribution for β-alanine and γ-ABA. The amino acid enantiomeric (d/l) ratios were also acquired in this study, and the values are all below 0.7, indicating some degree of terrestrial contamination. However, the d/l ratios are between 0.5 and 0.69, which is not too low to be recognized as complete terrestrial contamination. The possibility of an initial source of some indigenous extraterrestrial component cannot be completely ruled out. In fact, our study shows that the ordinary chondrites which have experienced higher temperature in their history have higher organic contents when compared with the carbonaceous chondrites. Therefore, it is possible that amino acids may be formed from gas-grain reactions after the meteorite parent body cooled to much lower temperatures or during the cooling process.

Acknowledgments— The research was funded by The Science and Technology Facilities Council (STFC) and The Royal Society. H-S. Chan is grateful for a Janet Watson Scholarship from The Department of Earth Science and Engineering at Imperial College London. The authors would like to thank the Reviewers D. Glavin and H. Yabuta for their helpful comments, which improved this publication.

Editorial Handling— Dr. Ian Franchi

Ancillary