The aim of this study was to investigate whether G. duodenalis, E. histolytica and Cryptosporidium sp. may have been present in the Near East region prior to the Roman period. The Near East is the region of the world where humans first created settlements, learned to farm and domesticate animals, and where the first large towns and cities developed (Bourke, Reference Bourke2018 ). As dysentery is more easily spread in environments with overcrowding, lack of organized sanitation and sewage systems, lack of understanding of how such diseases spread, and plenty of flies, we might expect the early cities of the Near East to have been well suited to disease outbreaks.

In medical texts from 2nd and 1st century millennium BCE Mesopotamia (ancient Iran and Iraq), the cuneiform word used to describe diarrhoea was sà si-sá. Diarrhoea is described in these texts as affecting infants and adults, and some texts describe incantations that they believed would help the sick person recover (Scurlock, Reference Scurlock2014 , 265; Steinert and Vacín, Reference Steinert, Vacín, Panayotov and Vacín2018 ). While these early written sources cannot allow us to differentiate the many causes of diarrhoea, they do encourage us to apply modern techniques to investigate which pathogens might have been involved.

One sample of sediment from the House of Ahiel cesspit was available for analysis, and 3 samples from different areas of the Armon ha-Natziv cesspit. A 1 g subsample of each was disaggregated using 0.5% trisodium phosphate solution, to form a suspension. This was passed through a stack of microsieves with mesh sizes 300, 160 and 20 μm to remove large soil particles, and the material that passed through the 20 μm sieve was used for ELISA analysis. This is because the cysts and oocysts of Entamoeba, Giardia and Cryptosporidium measure 5–19 μm in diameter (Garcia, Reference Garcia2016 ). The microsieves were thoroughly cleaned in an ultrasonicator bath with detergent between each sample. The sieved suspension was then centrifuged to concentrate the volume required for the ELISA plates, in the process concentrating the component of the sediment that should contain the protozoa if present. Following the manufacturer's instructions, a positive and a negative control were included in each microassay plate. A column of 8 wells was used for each sample. An ELISA plate reader (BioTek Synergy HT, Santa Clara, California, USA) was set to 450 nm and used to generate the absorbance values. Positive and negative results were allocated following the manufacturer recommended absorbance values, with more than 0.150 absorbance value being positive. This analysis was repeated in its entirety (using different sediment subsamples) on 3 separate dates over the course of a 12 month period to ensure reproducibility of the results.

The ‘House of Ahiel’ was a domestic building composed of 7 rooms, 4 of which form the main part of the building (Shiloh, Reference Shiloh1984 , 18 and Figs 20 and 25; Steiner, Reference Steiner2001 ). In room L789 a large locally carved block stone toilet seat was found ( Fig. 2B ) (Cahill et al., Reference Cahill, Reinhard, Tarler and Warnock1991 ), virtually identical in design to the Armon ha-Natziv toilet seat. The stone was positioned above a plastered cesspit and so seems to be positioned in its original place of use. The date of the construction of the House of Ahiel remains tentative at around the 8th century BCE, with some scholars suggesting an earlier date (Cahill, Reference Cahill, Vaughn and Killebrew2003 ). The destruction of the building is safely dated to 586 BCE, the Babylonian destruction of Jerusalem (Shiloh, Reference Shiloh1984 , 18). Microscopy of sediment from this cesspit by Karl Reinhard has identified the eggs of whipworm and Taenia sp. tapeworm (Cahill et al., Reference Cahill, Reinhard, Tarler and Warnock1991 ).

In 2019–2020 a salvage excavation by the Israel Antiquities Authority at Armon ha-Natziv (south Jerusalem, Fig. 1 ) exposed an estate which included a collection of ornamented architectural elements made of soft limestone including medium-sized ‘Proto-Aeolian’ stone capitals, fragments of lavish window frames and balustrades made of stylish columns. The level of workmanship in these capitals is of the highest standard known to date in the southern Levant during the Iron Age (Billig, Reference Billig2021 ; Billig et al., Reference Billig, Freud and Bocher2022 ). Based on ceramic typology, the site was dated to the mid-7th century BCE, probably the days of King Manasseh, who ruled over Judah for more than 50 years and was a client of the Assyrian empire (Gadot, Reference Gadot and Hagemeyer2022 ). According to Gadot the ornamental building that stood at the site should be understood as an Assyrian Bitanu. This suggestion is further supported by the evidence for an artificial garden (Langgut, Reference Langgut2022 ). It seems that the excavated area served as the garden of the estate and the actual building stood outside of the excavated lot. Within the estate garden a cubical stone object was found with a shallow curved surface for sitting, a large central hole for defecating and an adjacent hole likely for male urination ( Fig. 2A ). It was therefore interpreted as a stone toilet seat. It is thought that the toilet seat had fallen into the cesspit below after the floor support gave way, as stone slabs adjacent to the seat were steeply tilted downwards. Its dimensions are: 53 × 49 × 35 cm. Microscopy of sediment from this cesspit by Dafna Langgut has identified the eggs of whipworm, roundworm, Taenia sp. tapeworm and pinworm (Langgut, Reference Langgut2022 ).

Positive results from both latrine sediments were noted for G. duodenalis on the 3 dates the analysis was repeated. The House of Ahiel sample had between 2 out of 8 wells and 6 out of 8 wells positive on different analyses dates, while the 3 Armon ha-Natziv samples had all 8 wells positive each time the test was repeated. This might suggest that the Armon ha-Natziv cesspit sediment contained a higher concentration of Giardia antigen and cysts than did the House of Ahiel latrine, or that preservation of the antigen was better at the Armon ha-Natziv latrine. In contrast to the Giardia tests, the samples were negative for both Cryptosporidium sp. and E. histolytica. The Giardia ELISA plate values are given for each of the 3 analyses in Table 1 , and image of the plate is shown in Fig. 3 .

Discussion

The results presented here give what is currently the earliest known evidence for G. duodenalis (syn. G. lamblia, G. intestinalis, G. enterica) so far identified in a past population anywhere in the world. It has previously been identified in Roman period Turkey and also in Israel during the medieval and Ottoman periods (Mitchell et al., Reference Mitchell, Stern and Tepper2008; Yeh et al., Reference Yeh, Prag, Clamer, Humbert and Mitchell2015; Williams et al., Reference Williams, Arnold-Foster, Yeh, Ledger, Baeten, Poblome and Mitchell2017; Eskew et al., Reference Eskew, Ledger, Lloyd, Pyles, Gosker and Mitchell2019). In light of this, these results from Iron Age Jerusalem likely indicate the long-term presence of this parasite in the populations of the Near East.

Reliability of the result In view of the significance of this result, we should carefully explore the reliability of this analysis. The latrines were clearly identified as such by the presence of toilet seats. The samples were taken from the cesspit beneath each seat by the archaeologists excavating latrine, with each latrine excavated by different archaeologists some decades apart. The cesspit sediment from each latrine was found to contain intestinal helminth eggs on microscopy (Cahill et al., Reference Cahill, Reinhard, Tarler and Warnock1991; Langgut, Reference Langgut2022). They were no longer used after the destruction of Jerusalem by the Babylonians in 586 BCE. Therefore, it is highly unlikely that the sampled sediment was contaminated by the environmental conditions or by those excavating the site. The Techlab Giardia II ELISA kits use monoclonal and polyclonal antibodies to detect the cyst wall protein 1 (CWP1), which is a stable protein produced and released by encysting Giardia trophozoites (Boone et al., Reference Boone, Wilkins, Nash, Brandon, Macias, Jerris and Lyerly1999). No test is completely accurate all the time, so an understanding of the limitations of the tests used here is helpful. In a study using fresh stool microscopy, polymerase chain reaction and the Techlab Giardia II kit, sensitivity was 97% and specificity was 100% for the ELISA kit (Silva et al., Reference Silva, Pacheco, Martins, Menezes, Costa-Ribeiro, Ribeiro, Mattos, Oliveira, Soares and Teixeira2016). Another study based in 3 separate institutions that used fresh stool microscopy and ELISA found the Techlab Giardia II kit to have 91–100% sensitivity, and 97.8–100% specificity, depending on the institution (Boone et al., Reference Boone, Wilkins, Nash, Brandon, Macias, Jerris and Lyerly1999). In other words, the test may sometimes miss a true infection (perhaps if cyst concentrations are low), but a positive result is highly likely to be genuine. We do accept that it is theoretically possible that detection of coproantigens such as CWP1 could be confounded by cross-reactions with site-specific environmental antigens. However, our ability to cross-check with alternative molecular methods is outside the scope of our current investigation. The fact that we found samples from both Iron Age latrines to be positive on repeating the entire analysis process three times on separate dates is reassuring. On our last test run we left empty columns (plate columns 2, 4, 6, 8, 10) between the Iron Age samples (plate columns 3, 5, 7, 9) to ensure there was no contamination from 1 column to the next during the microassay plate washing required for various steps of the analysis, and the columns between the Iron Age samples gave a clear negative result (see Fig. 3). This again is reassuring.

Implications for our understanding of ancient populations of the Near East Giardia is a flagellated protozoan parasite that lives in the small intestine as a pear-shaped trophozoite measuring 9–20 μm in size, and as an oval infectious cyst that typically measures 8–12 μm. Recent assessment of the gene sequences of Giardia species complex has noted a number of distinct host-specific assemblages (Wielinga et al., Reference Wielinga, Williams, Monis and Thompson2023), but detection using ELISA is not able to differentiate each assemblage. Giardia is spread by the contamination of water or food with the feces of an infected person or non-human mammal. Trophozoites attach themselves to the lining of the intestine, which results in inflammation and damage to the epithelium and microvilli. Symptomatic infection by Giardia is termed giardiasis. Common symptoms include diarrhoea, abdominal cramps, malabsorption and weight loss. However, not all infections cause symptoms (Ryan et al., Reference Ryan, Hijjawi, Feng and Xiao2019; Adam, Reference Adam2021). Many individuals fully recover after an acute infection, but up to a third can experience chronic diseases such as post-infective irritable bowel, ocular pathology, arthritis, allergies and muscular complications. Most of those who die from Giardia are children, and chronic infection in this group can lead to stunted growth, impaired cognitive function and failure to thrive (Halliez and Buret, Reference Halliez and Buret2013). The fact that the sediment from both Iron Age cesspits was positive for Giardia would suggest that this parasite was endemic in the region of Jerusalem and the Kingdom of Judah during the 7th to early 6th century BCE. Since there was trade and military expeditions taking place across the Near East throughout this time period, we would expect such gastrointestinal infections to be spread easily by travellers. The many large and crowded towns and cities existing across the Near East by this time would have been fertile areas for the spread of such infections. While they did have toilets with cesspits across the region by the Iron Age, they were relatively rare and often only made for the elite. Towns were not planned and built with a sewerage network, flushing toilets had yet to be invented and the population had no understanding of existence of micro-organisms and how they can be spread (McMahon, Reference McMahon and Mitchell2015). Furthermore, the house fly (Musca domestica) is widespread in the Near East, and is well known for its ability to spread enteric pathogens that cause diarrhoea (Bidawid et al., Reference Bidawid, Edeson, Ibrahim and Matossian1978; Cohen et al., Reference Cohen, Green, Block, Slepon, Ambar, Wasserman and Levine1991). Therefore, it is probable that such flies contributed to the spread of diarrhoeal illness in the ancient Near East as well. It seems that at least some of those descriptions of diarrhoea in 2nd and 1st millennium BCE Mesopotamian medical texts (Scurlock, Reference Scurlock2014, 265; Steinert and Vacín, Reference Steinert, Vacín, Panayotov and Vacín2018) may well have included individuals suffering with giardiasis.